Error if onion payloads exceed max length on packet construction.
Ensure that if we call construct_onion_packet and friends where payloads are
too large for the allotted packet length, we'll fail to construct. Previously,
senders would happily construct invalid packets by array-shifting the final
node's HMAC out of the packet when adding an intermediate onion layer, causing
the receiver to error with "final payload provided for us as an intermediate
node."
Fix debug panic in onion utils on large custom TLVs or metadata.
We previously assumed that the final node's payload would be ~93 bytes, and had
code to ensure that the filler encoded after that payload is not all 0s. Now
with custom TLVs and metadata supported, the final node's payload may take up
the entire onion packet, so we can't assume that there are 64 bytes of filler
to check.
Matt Corallo [Wed, 6 Dec 2023 05:29:28 +0000 (05:29 +0000)]
Pre-calculate heap element scores (retaining RouteGraphNode size)
`RouteGraphNode` currently recalculates scores in its `Ord`
implementation, wasting time while sorting the main Dijkstra's
heap.
Further, some time ago, when implementing the `htlc_maximum_msat`
amount reduction while walking the graph, we added
`PathBuildingHop::was_processed`, looking up the source node in
`dist` each time we pop'ed an element off of the binary heap.
As a result, we now have a reference to our `PathBuildingHop` when
processing a best-node's channels, leading to several fields in
`RouteGraphNode` being entirely redundant.
Here we drop those fields, but add a pre-calculated score field,
as well as force a suboptimal `RouteGraphNode` layout, retaining
its existing 64 byte size.
Without the suboptimal layout, performance is very mixed, but with
it performance is mostly improved, by around 10% in most tests.
Matt Corallo [Wed, 6 Dec 2023 05:02:07 +0000 (05:02 +0000)]
Reorder `PathBuildingHop` fields somewhat
Given `PathBuildingHop` is now an even multiple of cache lines, we
can pick which fields "fall off" the cache line we have visible
when dealing with hops, which we do here.
Matt Corallo [Wed, 6 Dec 2023 06:02:37 +0000 (06:02 +0000)]
Make `find_route`'s `dist` map elements fit in 128 bytes
We'd previously aggressively cached elements in the
`PathBuildingHop` struct (and its sub-structs), which resulted in a
rather bloated size. This implied cache misses as we read from and
write to multiple cache lines during processing of a single
channel.
Here, we reduce caching in `DirectedChannelInfo`, fitting the
`(NodeId, PathBuildingHop)` tuple in exactly 128 bytes. While this
should fit in a single cache line, it sadly does not generally lie
in only two lines, as glibc returns large buffers from `malloc`
which are very well aligned, plus 16 bytes (for its own allocation
tracking). Thus, we try to avoid reading from the last 16 bytes of
a `PathBuildingHop`, but luckily that isn't super hard.
Note that here we make accessing
`DirectedChannelInfo::effective_capacity` somewhat slower, but
that's okay as its only ever done once per `DirectedChannelInfo`
anyway.
While our routing benchmarks are quite noisy, this appears to
result in between a 5% and 15% performance improvement in the
probabilistic scoring benchmarks.
Matt Corallo [Wed, 6 Dec 2023 03:54:28 +0000 (03:54 +0000)]
Make `CandidateRouteHop::PrivateHop::target_node_id` a reference
This avoids bloating `CandidateRouteHop` with a full 33-byte
node_id (and avoids repeated public key serialization when we do
multiple pathfinding passes).
Matt Corallo [Wed, 6 Dec 2023 17:47:00 +0000 (17:47 +0000)]
Simplify and make scoring calls in `TestRouter` more complete
`TestRouter` tries to make scoring calls that mimic what an actual
router would do, but the changes in f0ecc3ec73dcdb9303b1bd5ac687a36
failed to make scoring calls for private hints or if we take a
public hop for the last hop.
This fixes those regressions, though no tests currently depend on
this behavior.
Matt Corallo [Wed, 6 Dec 2023 01:17:48 +0000 (01:17 +0000)]
`#[inline]` `CandidateRouteHop` accessors
These are used in the performance-critical routing and scoring
operations, which may happen outside of our crate. Thus, we really
need to allow downstream crates to inline these accessors into
their code, which we do here.
Matt Corallo [Wed, 6 Dec 2023 01:13:33 +0000 (01:13 +0000)]
Privatise `CandidateRouteHop::short_channel_id` as its a footgun
Short channel "ID"s are not globally unique when they come from a
BOLT 11 route hint or a first hop (which can be an outbound SCID
alias). In those cases, its rather confusing that we have a
`short_channel_id` method which mixes them all together, and even
more confusing that we have a `CandidateHopId` which is not, in
fact returning a unique identifier.
In our routing logic this is mostly fine - the cost of a collision
isn't super high and we should still do just fine finding a route,
however the same can't be true for downstream users, as they may or
may not rely on the apparent guarantees.
Thus, here, we privatise the SCID and id accessors.
Matt Corallo [Wed, 6 Dec 2023 17:48:51 +0000 (17:48 +0000)]
Fix and re-enable the `remembers_historical_failures` test
f0ecc3ec73dcdb9303b1bd5ac687a361decce2dd introduced a regression in
the `remembers_historical_failures` test, and disabled it by simply
removing the `#[test]` annotation. This fixes the test and marks it
as a test again.
Elias Rohrer [Fri, 8 Dec 2023 11:29:38 +0000 (12:29 +0100)]
Manually download `bitcoind`/`electrs` for CI tests
Previously, we used the auto-download feature of the
`electrsd`/`bitcoind` crates. While convenient, they unnecessarily
introduced a lot of dependecies (`zip`, `zstd`, `time`, etc.) to our
test environment which needed pinning for the MSRV of 1.63.
Here, we introduce a new `no_download` config flag to the
`lightning-transaction-sync` crate allowing us to disable this
auto-download feature in CI, where we now opt to download the
corresponding binaries manually. We keep the default-auto-download as a
convenience feature for running tests locally though.
Jeffrey Czyz [Fri, 8 Dec 2023 04:44:58 +0000 (22:44 -0600)]
Return correct SendSuccess in OnionMessenger
When enqueuing a message for a node already awaiting a connection,
BufferedAwaitingConnection should be returned when a node is not yet
connected as a peer. However, it was only returned when the first
message was enqueued. Any messages enqueued after but before a
connection was established incorrectly returned Buffered.
Matt Corallo [Wed, 29 Nov 2023 06:02:46 +0000 (06:02 +0000)]
Immediately error in `close_channel_internal` if there is no chan
Previously, unfunded channels would be stored outside of
`PeerState::channel_by_id`, and thus if there is no channel when
we look in `PeerState::channel_by_id`, `close_channel_internal`
called `force_close_channel_with_peer` to hunt for unfunded
channels.
However, that is no longer the case, so the call is redundant, and
we can simply return an error instead.
Matt Corallo [Wed, 29 Nov 2023 18:11:30 +0000 (18:11 +0000)]
Move pre-funded-channel immediate shutdown logic to the right place
Because a `Funded` `Channel` cannot possibly be pre-funding, the
logic in `ChannelManager::close_channel_internal` to handle
pre-funding channels is in the wrong place.
Rather than being handled inside the `Funded` branch, it should be
in an `else` following it, handling either of the two
`ChannelPhases` outside of `Funded`.
Sadly, because of a previous control flow management `loop {}`, the
existing code will infinite loop, which is fixed here.
Matt Corallo [Thu, 30 Nov 2023 00:48:37 +0000 (00:48 +0000)]
Limit the scope of `get_funding_created_msg` to outbound channels
Since we no longer use `ChannelContext::get_funding_created_msg`,
it can be moved back into `UnfundedOutboundV1` channels only,
where it realistically belongs.
Matt Corallo [Thu, 30 Nov 2023 00:36:16 +0000 (00:36 +0000)]
Move to `Funded` after `funding_signed` rather than on funding
Previously, channels were stored in different maps in `PeerState`
based on whether the funding had been set, keeping the keys across
the maps consistent (pre-funding temporary_channel_ids vs
funding-outpoint-based channel_ids). However, channels are now
stored in a single `channel_by_id` map, making that point moot.
Instead, here, we convert the `ChannelPhase` state transition
boundary to "once we have a `ChannelMonitor`", which makes more
sense now, and was actually the original proposed boundary.
This also requires calling `signer_maybe_unblocked` on a pre-funded
outbound channel, but that nicely also lets us limit the scope of
`FundingCreated` message generation, which we do in the next
commit.
Wilmer Paulino [Tue, 5 Dec 2023 23:38:47 +0000 (15:38 -0800)]
Rename certain flags to align with dual funding
`FundingCreated` and `FundingSent` were mostly named after the
respective `funding_created` and `funding_sent` wire messages. They
include the signature for the initial commitment transaction when
opening a channel. With dual funding, these messages are no longer used,
and instead we rely on the existing `commitment_signed` to exchange
those signatures.
Jeffrey Czyz [Thu, 30 Nov 2023 03:30:15 +0000 (21:30 -0600)]
Test pending connection onion message buffering
Add tests for onion message buffering checking that messages are cleared
upon disconnection and timed out after MAX_TIMER_TICKS. Also, checks
that ConnectionNeeded events are generated.
Jeffrey Czyz [Thu, 16 Nov 2023 16:21:12 +0000 (10:21 -0600)]
Call OnionMessageHandler::timer_tick_occurred
lightning-background-processor processes events provided by the
PeerManager's OnionMessageHandler for when a connection is needed. If a
connection is not established in a reasonable amount of time, drop any
buffered onion messages by calling timer_tick_occurred.
Jeffrey Czyz [Thu, 16 Nov 2023 16:07:12 +0000 (10:07 -0600)]
Process OnionMessageHandler events in background
OnionMessageHandler implementations now also implement EventsProvider.
Update lightning-background-processor to also process any events the
PeerManager's OnionMessageHandler provides.
Jeffrey Czyz [Thu, 9 Nov 2023 21:58:24 +0000 (15:58 -0600)]
Drop buffered messages for timed out nodes
OnionMessenger buffers onion messages for nodes that are pending a
connection. To prevent DoS concerns, add a timer_tick_occurred method to
OnionMessageHandler so that buffered messages can be dropped. This will
be called in lightning-background-processor every 10 seconds.
Jeffrey Czyz [Thu, 9 Nov 2023 17:13:01 +0000 (11:13 -0600)]
Make OnionMessageHandler extend EventsProvider
An OnionMessageHandler may buffer messages that can't be sent because
the recipient is not a peer. Have the trait extend EventsProvider so
that implementation so that an Event::ConnectionNeeded can be generated
for any nodes that fall into this category. Also, implement
EventsProvider for OnionMessenger and IgnoringMessageHandler.
Jeffrey Czyz [Thu, 9 Nov 2023 17:10:23 +0000 (11:10 -0600)]
Add Event::ConnectionNeeded for onion messages
A MessageRouter may be unable to find a complete path to an onion
message's destination. This could because no such path exists or any
needs on a potential path don't support onion messages. Add an event
that indicates a connection with a node is needed in order to send the
message.
Jeffrey Czyz [Tue, 14 Nov 2023 23:08:26 +0000 (17:08 -0600)]
Return socket addresses from DefaultMessageRouter
When there isn't a direct connection with the Destination of an
OnionMessage, look up socket addresses from the NetworkGraph. This is
used to signal to OnionMessenger that a direct connection is needed to
send the message.
Jeffrey Czyz [Tue, 14 Nov 2023 21:30:17 +0000 (15:30 -0600)]
Add Option<Vec<SocketAddress>> to OnionMessagePath
MessageRouter::find_path is given a Destination to reach via a set of
peers. If a path cannot be found, it may return a partial path such that
OnionMessenger can signal a direct connection to the first node in the
path is needed. Include a list of socket addresses in the returned
OnionMessagePath to allow OnionMessenger to know how to connect to the
node.
This allows DefaultMessageRouter to use its NetworkGraph to return
socket addresses for gossiped nodes.
Jeffrey Czyz [Tue, 14 Nov 2023 21:05:05 +0000 (15:05 -0600)]
Add NetworkGraph reference to DefaultMessageRouter
When buffering onion messages for a node that is not connected as a
peer, it's possible that the node does not exist. Include a NetworkGraph
reference in DefaultMessageRouter so that it can be used to check if the
node actually exists. Otherwise, an malicious node may send an onion
message where the reply path's introduction node doesn't exist. This
would result in buffering messages that may never be delivered.
Jeffrey Czyz [Tue, 7 Nov 2023 14:17:46 +0000 (08:17 -0600)]
Buffer onion messages requiring a connection
MessageRouter::find_path returns a path to use when sending an onion
message. If the first node on the path is not connected or does not
support onion messages, sending will fail with InvalidFirstHop. Instead
of failing outright, buffer the message for later sending once the first
node is a connected peer.
Jeffrey Czyz [Wed, 15 Nov 2023 23:26:45 +0000 (17:26 -0600)]
Destination in OnionMessenger::send_onion_message
OnionMessenger::send_onion_message takes an OnionMessagePath. This isn't
very useful as it requires finding a path manually. Instead, have the
method take a Destination and use OnionMessenger's MessageRouter to
construct the path. Later, this will allow for buffering messages where
the first node in the path isn't a direct connection.
Jeffrey Czyz [Mon, 6 Nov 2023 22:53:07 +0000 (16:53 -0600)]
Use a message buffer abstraction in OnionMessenger
Onion messages are buffered for sending to the next node. Since the
network has limited adoption, connecting directly to a peer may be
necessary. Add an OnionMessageBuffer abstraction that can differentiate
between connected peers and those are pending a connection. This allows
for buffering messages before a connection is established and applying
different buffer policies for peers yet to be connected.
Matt Corallo [Wed, 29 Nov 2023 23:25:04 +0000 (23:25 +0000)]
Allow users to accept skimmed fees in calling `peel_payment_onion`
LSP users who wish to use `peel_payment_onion` to understand if
they'd accept an HTLC prior to receit should be able to check the
skimmed fees just like they would for full payment receipt. Thus,
we need to expose the fee-skimming acceptance bool to
`peel_payment_onion`, which we do here, in addition to some doc
cleanups.
Matt Corallo [Wed, 29 Nov 2023 23:23:34 +0000 (23:23 +0000)]
Flesh out docs on `PendingHTLCInfo`
Now that `PendingHTLCInfo` is public, its docs should be meaningful
to developers not working directly on LDK, and thus needs
substantially more information than it previously had.
Matt Corallo [Wed, 29 Nov 2023 23:11:02 +0000 (23:11 +0000)]
Flesh out docs on `PendingHTLCRouting`
Now that `PendingHTLCRouting` is public, its docs should be
meaningful to developers not working directly on LDK, and thus
needs substantially more information than it previously had.
Wilmer Paulino [Tue, 24 Oct 2023 23:54:20 +0000 (16:54 -0700)]
Refactor ChannelState to decouple state flags from states
Previously, our `ChannelState` contained bits for both states and flags.
To make matters worse, some of the flags could apply to multiple states.
This led to its API being very cumbersome, having to apply masks in most
scenarios to check for certain states. As LDK grows and more features
are added requiring more states/flags, the need for a simpler API
arises.
This refactor aims to improve this by decoupling the state flags from
the `ChannelState` enum. Each state that requires flags will now have
its own flags type, to ensure flags can only be applied to their
intended state. All of this is done while maintaining backwards and
forwards compatibility.
jbesraa [Sat, 26 Aug 2023 15:06:45 +0000 (18:06 +0300)]
Add `source` and `target` fn's to `CandidateRouteHop`
We add `CandidateRouteHop::source` and
`CandidateRouteHop::source` functions to point
to current and next hops in route respectively.
As we have now `source` and `target`
available in `CandidateRouteHop` we also
remove `CandidateRouteHop::id` inputs
so now they are consumed from `self.target`
and `self.source` functions.
In the `add_entry` macro we also remove `source`
and `target` arguments in favor of `candidate`
of type `CandidateRouteHop` that holds the
needed info.
jbesraa [Tue, 5 Dec 2023 18:56:51 +0000 (20:56 +0200)]
Add `outbound` flag to `DirectedChannelInfo`
If `outband` flag is set to true then `ChannelInfo::node_one`
is forwarding a payment to target `ChannelInfo::node_two`.
If `outband` flag is set to false then `ChannelInfo::node_two`
is forwarding a payment to target `ChannelInfo::node_one`.
Matt Corallo [Tue, 28 Nov 2023 23:19:39 +0000 (23:19 +0000)]
Provide inbound HTLC preimages to the `EcdsaChannelSigner`
The VLS signer has a desire to see preimages for resolved forwarded
HTLCs when they are first claimed by us, even if that claim was for
the inbound edge (where claiming strictly increases our balance).
Luckily, providing that information is rather trivial, which we do
here.
Elias Rohrer [Mon, 4 Dec 2023 14:39:13 +0000 (15:39 +0100)]
Log additional details when ignoring first hops
Users are often confused when we fail to find a route due to some
requirements on the first hop are not being met. While we now take note
and log such candidates, we still previously required users to check
additional details to figure out why exactly the router refused to route
over a particular first hop.
Here, we add additional TRACE logging, in particular for
`ChannelDetails::next_outbound_htlc_limit_msat` and
`ChannelDetails::next_outbound_htlc_minimum_msat` when they are
relevant.
Matt Corallo [Sun, 3 Dec 2023 19:51:24 +0000 (19:51 +0000)]
Remove redundant locking when creating `WithChannelMonitor`
The `WithChannelMonitor` log decorator redundantly locks the
`ChannelMonitor` inner mutex, which we fix here, as well as add a
new constructor which avoids locking at all if an inner mutex lock
is already readily available.
Matt Corallo [Sat, 2 Dec 2023 18:45:03 +0000 (18:45 +0000)]
Drop `log_bytes!()` calls on `Display` types in `ChannelManager`
df3ab2ee2753e7f9ec02ddf1c8a51db77c50e35d was rebased one too many
times and ended up reverting some of the `log_bytes!()` removals
around types which now implement `Display` in `ChannelManager`.
This commit removes those, as well as one additoinal excess macro
which slipped in somewhere else.
Matt Corallo [Sat, 2 Dec 2023 17:54:18 +0000 (17:54 +0000)]
Correct lock drop'ing in `ChainMonitor::update_channel`
e21a500668179c7084e2df5cb75019810eb03cbe cleaned up the error
handling in `ChainMonitor::update_channel` a bit, but accidentally
replaced the deliberate panic with a hang. This commit ensures we
properly drop the monitors read lock before taking a write lock.
Matt Corallo [Sat, 2 Dec 2023 17:54:09 +0000 (17:54 +0000)]
Add some missing `WithContext` wrappers in `peer_handler.rs`
df3ab2ee2753e7f9ec02ddf1c8a51db77c50e35d added `WithContext`
wrappers in most logs in `peer_handler.rs`, but due to long-term
rebasing it missed a few, which we add here.
Matt Corallo [Sat, 2 Dec 2023 04:18:30 +0000 (04:18 +0000)]
Lean on `Logger` wrapping in `ChannelMonitor` rather than in channel
Now that `ChannelMonitor` is careful about wrapping `Logger`s at
the edge, there's no need to use `WithChannelMonitor` in a few
cases in `channel.rs` and one in `channelmanager.rs`.
Matt Corallo [Sun, 3 Dec 2023 20:03:51 +0000 (20:03 +0000)]
Ensure `ChannelMonitor` `Logger`s are always wrapped with metadata
In order to ensure log lines generated by `ChannelMonitor` always
have a counterparty and channel ID entry, this consistently wraps
`Logger`s in a decorator in all `pub(X)` `ChannelMonitor` functions,
removing `pub` markings on `ChannelMonitorImpl` methods that aren't
actually publicly reachable anyway.
This also lets us clean up the `Logger` types in various
`ChannelMonitor` methods.
Matt Corallo [Sun, 3 Dec 2023 19:09:32 +0000 (19:09 +0000)]
Doc that `Record::node_id` may be missing even with `channel_id`
There are various place where we log something related to a channel
but fail to fill in the channel's counterparty information. This is
somewhat surprising, given channel counterparty information is
always known, but simply is sometimes not readily accessible to LDK
when a log is printed.
Matt Corallo [Sat, 2 Dec 2023 19:13:02 +0000 (19:13 +0000)]
Marginally optimize test logging
973636bd2ab2ba35fb8b9703f1d5d0e72f069cdc introduced a new `HashMap`
in the `TestLogger` but then did lookups by iterating the entire
map. This fixes that, and also takes this opportunity to stop
allocating new `String`s for the module to store each log entry in
the `TestLogger`