Matt Corallo [Wed, 12 Jun 2024 20:49:07 +0000 (20:49 +0000)]
Track the cp `node_id` which originated an HTLC in the `HTLCSource`
Because we track pending `ChannelMonitorUpdate`s per-peer, we
really need to know what peer an HTLC came from when we go to claim
it on-chain, allowing us to look up any blocked actions in the
`PeerState`. While we could do this by moving the blocked actions
to some global "closed-channel update queue", its simpler to do it
this way.
While this trades off `ChannelMonitorUpdate` size somewhat (the
`HTLCSource` is included in many of them), which we should be
sensitive to, it will also allow us to (eventually) remove the
SCID -> peer + channel_id lookups we do when claiming or failing
today, which are somewhat annoying to deal with.
Matt Corallo [Tue, 18 Jun 2024 17:24:21 +0000 (17:24 +0000)]
Add skipping variants to `impl_writeable_tlv_based_enum_upgradable`
In some cases, we have variants of an enum serialized using
`impl_writeable_tlv_based_enum_upgradable` which we don't want to
write/read. Here we add support for such variants by writing them
using the (odd) type 255 without any contents and using
`MaybeReadable` to decline to read them.
AsyncPaymentsMessageHandler trait for OnionMessenger
Add a trait for handling async payments messages to OnionMessenger. This allows
users to either provide their own custom handling for async payments messages
or rely on a version provided by LDK.
Willem Van Lint [Tue, 11 Jun 2024 23:05:57 +0000 (16:05 -0700)]
Implement non-strict forwarding
This change implements non-strict forwarding, allowing the node to
forward an HTLC along a channel other than the one specified
by short_channel_id in the onion message, so long as the receiver has
the same node public key intended by short_channel_id
([BOLT](https://github.com/lightning/bolts/blob/57ce4b1e05c996fa649f00dc13521f6d496a288f/04-onion-routing.md#non-strict-forwarding)).
This can improve payment reliability when there are multiple channels
with the same peer e.g. when outbound liquidity is replenished by
opening a new channel.
The implemented forwarding strategy now chooses the channel with the
lowest outbound liquidity that can forward an HTLC to maximize the
probability of being able to successfully forward a subsequent HTLC.
G8XSU [Fri, 14 Jun 2024 23:56:36 +0000 (16:56 -0700)]
Optimize ChannelMonitor persistence on block connections.
Currently, every block connection triggers the persistence of all
ChannelMonitors with an updated best_block. This approach poses
challenges for large node operators managing thousands of channels.
Furthermore, it leads to a thundering herd problem
(https://en.wikipedia.org/wiki/Thundering_herd_problem), overwhelming
the storage with simultaneous requests.
To address this issue, we now persist ChannelMonitors at a
regular cadence, spreading their persistence across blocks to
mitigate spikes in write operations.
Outcome: After doing this, Ldk's IO footprint should be reduced
by ~50 times. The processing time required to sync each block
will be significantly reduced, particularly for nodes with 1000s
of channels, as write latency plays a significant role in this process.
As a result, the Node/ChainMonitor will be blocked for a shorter
duration, leading to further efficiency gains.
Jeffrey Czyz [Mon, 17 Jun 2024 19:41:32 +0000 (14:41 -0500)]
Relax channel count check for unannounced nodes
When creating blinded paths, introduction nodes are limited to peers
with at least three channels to prevent easily guessing the recipient.
Relax this check when the recipient is unannounced since they won't be
in the NetworkGraph.
Jeffrey Czyz [Thu, 13 Jun 2024 22:15:19 +0000 (17:15 -0500)]
Advance self blinded payment paths
DefaultRouter will ignore blinded paths where the sender is the
introduction node. Similar to message paths, advance such paths by one
so that payments may be sent to them.
Jeffrey Czyz [Fri, 14 Jun 2024 21:57:41 +0000 (16:57 -0500)]
Add advance_path_by_one for blinded payment paths
Similar to blinded paths for onion messages, if given a blinded payment
path where we are the introduction node, the path must be advanced by
one in order to use it.
Jeffrey Czyz [Fri, 14 Jun 2024 14:04:47 +0000 (09:04 -0500)]
Don't modify path when advance_path_by_one errors
When using advance_path_by_one when we are the introduction node, any
error will result having the first hop of the input blinded path
removed. Instead, only remove the first hop on success. Otherwise, the
path will be invalid.
Jeffrey Czyz [Wed, 12 Jun 2024 21:54:19 +0000 (16:54 -0500)]
Blinded payments to unannounced introduction node
When creating blinded paths for receiving onion payments, allow using
the recipient's only peer as the introduction node when the recipient is
unannounced. This allows for sending payments without going through an
intermediary, which is useful for testing or when only connected to an
LSP with an empty NetworkGraph.
Jeffrey Czyz [Wed, 12 Jun 2024 15:58:18 +0000 (10:58 -0500)]
BlindedPath with unannounced introduction node
When creating blinded paths for receiving onion messages, allow using
the recipient's only peer as the introduction node when the recipient is
unannounced. This allows for sending messages without going through an
intermediary, which is useful for testing or when only connected to an
LSP with an empty NetworkGraph.
Jeffrey Czyz [Tue, 21 May 2024 23:47:32 +0000 (18:47 -0500)]
Expose send_payment_for_bolt12_invoice
UserConfig::manually_handle_bolt12_invoices allows deferring payment of
BOLT12 invoices by generating Event::InvoiceReceived. Expose
ChannelManager::send_payment_for_bolt12_invoice to allow users to pay
the Bolt12Invoice included in the event. While the event contains the
PaymentId for reference, that parameter is now removed from the method
in favor of extracting the PaymentId from the invoice's payer_metadata.
Jeffrey Czyz [Tue, 21 May 2024 21:58:10 +0000 (16:58 -0500)]
Add UserConfig::manually_handle_bolt12_invoices
BOLT12 invoices are automatically paid once they have been verified.
Users may want to manually pay them by first performing additional
checks. Add a manually_handle_bolt12_invoices configuration option that
when set generates an Event::InvoiceReceived instead of paying the
invoice.
Add a builder for creating static invoices for an offer. Building produces a
semantically valid static invoice for the offer, which can then be signed with
the key associated with the offer's signing pubkey.
Jeffrey Czyz [Tue, 21 May 2024 19:33:56 +0000 (14:33 -0500)]
Add an InvoiceReceived event
Some users may want to handle a Bolt12Invoice asynchronously, either in
a different process or by first performing additional verification
before paying the invoice. Add an InvoiceReceived event to facilitate
this.
Matt Corallo [Wed, 12 Jun 2024 14:06:54 +0000 (14:06 +0000)]
Drop unnecessary key derivation in route tests
In our route tests we need some "random" bytes for the router to
randomize amounts using. We generate this by building an actual
`KeysManager` and then deriving some random bytes using the
`EntropySource` trait. However, `get_route` (what we're normally
testing) doesn't actually use the random bytes, and even if it did,
using a `KeysManager` is just a fancy way of creating a constant,
so there's really no reason to do all the fancy crypto.
Instead, here, we change our routing tests and benchmarks to simply
use `[42; 32]` as the "random" bytes.
Matt Corallo [Sun, 10 Dec 2023 18:23:17 +0000 (18:23 +0000)]
Use a real (probing-generated) scorer in benchmarks
Until now, our routing benchmarks used a synthetic scorer,
generated by scoring random paths to build up some history. This is
pretty far removed from real-world routing conditions, as
alternative paths generally have no scoring information and even
the paths we do take have only one or two past scoring results.
Instead, we fetch a static serialized scorer, generated using
minutely probes. This means future changes to the scorer's data may
be harder to benchmark, but makes for substantially more realistic
benchmarks for changes which don't impact the serialized state.
Define an interface for BOLT 12 static invoice messages. The underlying
format consists of the original bytes and the parsed contents.
The bytes are later needed for serialization. This is because it must
mirror all the offer TLV records, including unknown ones, which aren't
represented in the contents.
Matt Corallo [Tue, 11 Jun 2024 01:39:48 +0000 (01:39 +0000)]
Unblock channels awaiting monitor update based on `ChanMan` queue
When we have `ChannelMonitorUpdate`s which are completing both
synchronously and asynchronously, we need to consider a channel as
unblocked based on the `ChannelManager` monitor update queue,
rather than by checking the `update_id`s.
Consider the case where a channel is updated, leading to a
`ChannelMonitorUpdate` which completes asynchronously. The update
completes, but prior to the `ChannelManager` receiving the
`MonitorEvent::Completed` it generates a further
`ChannelMonitorUpdate`. This second update completes synchronously.
As a result, when the `MonitorEvent` is processed, the event's
`monitor_update_id` is the first update, but there are no updates
queued and the channel should be free to return to be unblocked.
Here we fix this by looking only at the `ChannelManager` update
queue, rather than the update_id of the `MonitorEvent`.
While we don't anticipate many users having both synchronous and
asynchronous persists in the same application, there isn't much
cost to supporting it, which we do here.
Matt Corallo [Mon, 15 Apr 2024 19:25:05 +0000 (19:25 +0000)]
Fuzz reloading with a stale monitor in chanmon_consistency
Now that we are gearing up to support fully async monitor storage,
we really need to fuzz monitor updates not completing before a
reload, which we do here in the `chanmon_consistency` fuzzer.
While there are more parts to async monitor updating that we need
to fuzz, this at least gets us started by having basic async
restart cases handled. In the future, we should extend this to make
sure some basic properties (eg claim/balance consistency) remain
true through `chanmon_consistency` runs.
Alec Chen [Thu, 30 May 2024 22:09:56 +0000 (17:09 -0500)]
Get per commitment point everywhere else with HolderCommitmentPoint
This includes when building TxCreationKeys, as well as for open_channel
and accept_channel messages. Note: this is only for places where we are
retrieving the current per commitment point, which excludes
channel_reestablish.
Matt Corallo [Sun, 5 May 2024 23:22:23 +0000 (23:22 +0000)]
Force-close channels if their feerate gets stale without any update
For quite some time, LDK has force-closed channels if the peer
sends us a feerate update which is below our `FeeEstimator`'s
concept of a channel lower-bound. This is intended to ensure that
channel feerates are always sufficient to get our commitment
transaction confirmed on-chain if we do need to force-close.
However, we've never checked our channel feerate regularly - if a
peer is offline (or just uninterested in updating the channel
feerate) and the prevailing feerates on-chain go up, we'll simply
ignore it and allow our commitment transaction to sit around with a
feerate too low to get confirmed.
Here we rectify this oversight by force-closing channels with stale
feerates, checking after each block. However, because fee
estimators are often buggy and force-closures piss off users, we
only do so rather conservatively. Specifically, we only force-close
if a channel's feerate is below the minimum `FeeEstimator`-provided
minimum across the last day.
Further, because fee estimators are often especially buggy on
startup (and because peers haven't had a chance to update the
channel feerates yet), we don't force-close channels until we have
a full day of feerate lower-bound history.
This should reduce the incidence of force-closures substantially,
but it is expected this will still increase force-closures somewhat
substantially depending on the users' `FeeEstimator`.
Matt Corallo [Fri, 31 May 2024 17:42:49 +0000 (17:42 +0000)]
Skip fee reads in `full_stack_target` when connecting many blocks
When we connect 100 blocks in a row, requiring the fuzz input to
contain 100 fee estimator results is uneccessary, so add a bool
that lets us skip those reads.
Matt Corallo [Sun, 5 May 2024 23:21:13 +0000 (23:21 +0000)]
Add a new `ClosureReason::PeerFeerateTooLow`
Closure due to feerate disagreements are a specific closure reason
which admins can understand and tune their config (in the form of
their `FeeEstimator`) to avoid, so having a separate
`ClosureReason` for it is useful.
Matt Corallo [Sun, 5 May 2024 23:52:08 +0000 (23:52 +0000)]
Use `ChannelError::close` constructor when building a close variant
In the next commit we'll add a second field to
`ChannelError::Close` so here we prep by converting existing calls
to the constructor function, which is almost a full-file sed.