/// `ScoreUpdate` is used to update the scorer's internal state after a payment attempt.
pub trait ScoreUpdate {
/// Handles updating channel penalties after failing to route through a channel.
- fn payment_path_failed(&mut self, path: &Path, short_channel_id: u64);
+ fn payment_path_failed(&mut self, path: &Path, short_channel_id: u64, duration_since_epoch: Duration);
/// Handles updating channel penalties after successfully routing along a path.
- fn payment_path_successful(&mut self, path: &Path);
+ fn payment_path_successful(&mut self, path: &Path, duration_since_epoch: Duration);
/// Handles updating channel penalties after a probe over the given path failed.
- fn probe_failed(&mut self, path: &Path, short_channel_id: u64);
+ fn probe_failed(&mut self, path: &Path, short_channel_id: u64, duration_since_epoch: Duration);
/// Handles updating channel penalties after a probe over the given path succeeded.
- fn probe_successful(&mut self, path: &Path);
+ fn probe_successful(&mut self, path: &Path, duration_since_epoch: Duration);
}
/// A trait which can both lookup and update routing channel penalty scores.
#[cfg(not(c_bindings))]
impl<S: ScoreUpdate, T: DerefMut<Target=S>> ScoreUpdate for T {
- fn payment_path_failed(&mut self, path: &Path, short_channel_id: u64) {
- self.deref_mut().payment_path_failed(path, short_channel_id)
+ fn payment_path_failed(&mut self, path: &Path, short_channel_id: u64, duration_since_epoch: Duration) {
+ self.deref_mut().payment_path_failed(path, short_channel_id, duration_since_epoch)
}
- fn payment_path_successful(&mut self, path: &Path) {
- self.deref_mut().payment_path_successful(path)
+ fn payment_path_successful(&mut self, path: &Path, duration_since_epoch: Duration) {
+ self.deref_mut().payment_path_successful(path, duration_since_epoch)
}
- fn probe_failed(&mut self, path: &Path, short_channel_id: u64) {
- self.deref_mut().probe_failed(path, short_channel_id)
+ fn probe_failed(&mut self, path: &Path, short_channel_id: u64, duration_since_epoch: Duration) {
+ self.deref_mut().probe_failed(path, short_channel_id, duration_since_epoch)
}
- fn probe_successful(&mut self, path: &Path) {
- self.deref_mut().probe_successful(path)
+ fn probe_successful(&mut self, path: &Path, duration_since_epoch: Duration) {
+ self.deref_mut().probe_successful(path, duration_since_epoch)
}
}
} }
#[cfg(c_bindings)]
impl<'a, T: Score> ScoreUpdate for MultiThreadedScoreLockWrite<'a, T> {
- fn payment_path_failed(&mut self, path: &Path, short_channel_id: u64) {
- self.0.payment_path_failed(path, short_channel_id)
+ fn payment_path_failed(&mut self, path: &Path, short_channel_id: u64, duration_since_epoch: Duration) {
+ self.0.payment_path_failed(path, short_channel_id, duration_since_epoch)
}
- fn payment_path_successful(&mut self, path: &Path) {
- self.0.payment_path_successful(path)
+ fn payment_path_successful(&mut self, path: &Path, duration_since_epoch: Duration) {
+ self.0.payment_path_successful(path, duration_since_epoch)
}
- fn probe_failed(&mut self, path: &Path, short_channel_id: u64) {
- self.0.probe_failed(path, short_channel_id)
+ fn probe_failed(&mut self, path: &Path, short_channel_id: u64, duration_since_epoch: Duration) {
+ self.0.probe_failed(path, short_channel_id, duration_since_epoch)
}
- fn probe_successful(&mut self, path: &Path) {
- self.0.probe_successful(path)
+ fn probe_successful(&mut self, path: &Path, duration_since_epoch: Duration) {
+ self.0.probe_successful(path, duration_since_epoch)
}
}
}
impl ScoreUpdate for FixedPenaltyScorer {
- fn payment_path_failed(&mut self, _path: &Path, _short_channel_id: u64) {}
+ fn payment_path_failed(&mut self, _path: &Path, _short_channel_id: u64, _duration_since_epoch: Duration) {}
- fn payment_path_successful(&mut self, _path: &Path) {}
+ fn payment_path_successful(&mut self, _path: &Path, _duration_since_epoch: Duration) {}
- fn probe_failed(&mut self, _path: &Path, _short_channel_id: u64) {}
+ fn probe_failed(&mut self, _path: &Path, _short_channel_id: u64, _duration_since_epoch: Duration) {}
- fn probe_successful(&mut self, _path: &Path) {}
+ fn probe_successful(&mut self, _path: &Path, _duration_since_epoch: Duration) {}
}
impl Writeable for FixedPenaltyScorer {
}
impl<G: Deref<Target = NetworkGraph<L>>, L: Deref, T: Time> ScoreUpdate for ProbabilisticScorerUsingTime<G, L, T> where L::Target: Logger {
- fn payment_path_failed(&mut self, path: &Path, short_channel_id: u64) {
+ fn payment_path_failed(&mut self, path: &Path, short_channel_id: u64, _duration_since_epoch: Duration) {
let amount_msat = path.final_value_msat();
log_trace!(self.logger, "Scoring path through to SCID {} as having failed at {} msat", short_channel_id, amount_msat);
let network_graph = self.network_graph.read_only();
}
}
- fn payment_path_successful(&mut self, path: &Path) {
+ fn payment_path_successful(&mut self, path: &Path, _duration_since_epoch: Duration) {
let amount_msat = path.final_value_msat();
log_trace!(self.logger, "Scoring path through SCID {} as having succeeded at {} msat.",
path.hops.split_last().map(|(hop, _)| hop.short_channel_id).unwrap_or(0), amount_msat);
}
}
- fn probe_failed(&mut self, path: &Path, short_channel_id: u64) {
- self.payment_path_failed(path, short_channel_id)
+ fn probe_failed(&mut self, path: &Path, short_channel_id: u64, duration_since_epoch: Duration) {
+ self.payment_path_failed(path, short_channel_id, duration_since_epoch)
}
- fn probe_successful(&mut self, path: &Path) {
- self.payment_path_failed(path, u64::max_value())
+ fn probe_successful(&mut self, path: &Path, duration_since_epoch: Duration) {
+ self.payment_path_failed(path, u64::max_value(), duration_since_epoch)
}
}
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 301);
- scorer.payment_path_failed(&failed_path, 41);
+ scorer.payment_path_failed(&failed_path, 41, Duration::ZERO);
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 301);
- scorer.payment_path_successful(&successful_path);
+ scorer.payment_path_successful(&successful_path, Duration::ZERO);
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 301);
}
let usage = ChannelUsage { amount_msat: 750, ..usage };
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 602);
- scorer.payment_path_failed(&path, 43);
+ scorer.payment_path_failed(&path, 43, Duration::ZERO);
let usage = ChannelUsage { amount_msat: 250, ..usage };
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 0);
let usage = ChannelUsage { amount_msat: 750, ..usage };
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 602);
- scorer.payment_path_failed(&path, 42);
+ scorer.payment_path_failed(&path, 42, Duration::ZERO);
let usage = ChannelUsage { amount_msat: 250, ..usage };
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 300);
};
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 128);
- scorer.payment_path_failed(&Path { hops: path, blinded_tail: None }, 43);
+ scorer.payment_path_failed(&Path { hops: path, blinded_tail: None }, 43, Duration::ZERO);
let channel = network_graph.read_only().channel(42).unwrap().to_owned();
let (info, _) = channel.as_directed_from(&node_a).unwrap();
assert_eq!(scorer.channel_penalty_msat(&candidate_42, usage, ¶ms), 128);
assert_eq!(scorer.channel_penalty_msat(&candidate_43, usage, ¶ms), 128);
- scorer.payment_path_successful(&payment_path_for_amount(500));
+ scorer.payment_path_successful(&payment_path_for_amount(500), Duration::ZERO);
assert_eq!(scorer.channel_penalty_msat(&candidate_41, usage, ¶ms), 128);
assert_eq!(scorer.channel_penalty_msat(&candidate_42, usage, ¶ms), 300);
let usage = ChannelUsage { amount_msat: 1_023, ..usage };
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 2_000);
- scorer.payment_path_failed(&payment_path_for_amount(768), 42);
- scorer.payment_path_failed(&payment_path_for_amount(128), 43);
+ scorer.payment_path_failed(&payment_path_for_amount(768), 42, Duration::ZERO);
+ scorer.payment_path_failed(&payment_path_for_amount(128), 43, Duration::ZERO);
// Initial penalties
let usage = ChannelUsage { amount_msat: 128, ..usage };
};
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 125);
- scorer.payment_path_failed(&payment_path_for_amount(512), 42);
+ scorer.payment_path_failed(&payment_path_for_amount(512), 42, Duration::ZERO);
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 281);
// An unchecked right shift 64 bits or more in DirectedChannelLiquidity::decayed_offset_msat
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 300);
// More knowledge gives higher confidence (256, 768), meaning a lower penalty.
- scorer.payment_path_failed(&payment_path_for_amount(768), 42);
- scorer.payment_path_failed(&payment_path_for_amount(256), 43);
+ scorer.payment_path_failed(&payment_path_for_amount(768), 42, Duration::ZERO);
+ scorer.payment_path_failed(&payment_path_for_amount(256), 43, Duration::ZERO);
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 281);
// Decaying knowledge gives less confidence (128, 896), meaning a higher penalty.
// Reducing the upper bound gives more confidence (128, 832) that the payment amount (512)
// is closer to the upper bound, meaning a higher penalty.
- scorer.payment_path_successful(&payment_path_for_amount(64));
+ scorer.payment_path_successful(&payment_path_for_amount(64), Duration::from_secs(10));
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 331);
// Increasing the lower bound gives more confidence (256, 832) that the payment amount (512)
// is closer to the lower bound, meaning a lower penalty.
- scorer.payment_path_failed(&payment_path_for_amount(256), 43);
+ scorer.payment_path_failed(&payment_path_for_amount(256), 43, Duration::from_secs(10));
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 245);
// Further decaying affects the lower bound more than the upper bound (128, 928).
effective_capacity: EffectiveCapacity::Total { capacity_msat: 1_000, htlc_maximum_msat: 1_000 },
};
- scorer.payment_path_failed(&payment_path_for_amount(500), 42);
+ scorer.payment_path_failed(&payment_path_for_amount(500), 42, Duration::ZERO);
let channel = network_graph.read_only().channel(42).unwrap().to_owned();
let (info, _) = channel.as_directed_from(&source).unwrap();
let candidate = CandidateRouteHop::PublicHop {
SinceEpoch::advance(Duration::from_secs(10));
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 473);
- scorer.payment_path_failed(&payment_path_for_amount(250), 43);
+ scorer.payment_path_failed(&payment_path_for_amount(250), 43, Duration::from_secs(10));
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 300);
let mut serialized_scorer = Vec::new();
effective_capacity: EffectiveCapacity::Total { capacity_msat: 1_000, htlc_maximum_msat: 1_000 },
};
- scorer.payment_path_failed(&payment_path_for_amount(500), 42);
+ scorer.payment_path_failed(&payment_path_for_amount(500), 42, Duration::ZERO);
let channel = network_graph.read_only().channel(42).unwrap().to_owned();
let (info, _) = channel.as_directed_from(&source).unwrap();
let candidate = CandidateRouteHop::PublicHop {
<ProbabilisticScorer>::read(&mut serialized_scorer, (decay_params, &network_graph, &logger)).unwrap();
assert_eq!(deserialized_scorer.channel_penalty_msat(&candidate, usage, ¶ms), 473);
- scorer.payment_path_failed(&payment_path_for_amount(250), 43);
+ scorer.payment_path_failed(&payment_path_for_amount(250), 43, Duration::from_secs(10));
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 300);
SinceEpoch::advance(Duration::from_secs(10));
assert_eq!(scorer.historical_estimated_payment_success_probability(42, &target, 42, ¶ms),
None);
- scorer.payment_path_failed(&payment_path_for_amount(1), 42);
+ scorer.payment_path_failed(&payment_path_for_amount(1), 42, Duration::ZERO);
{
let network_graph = network_graph.read_only();
let channel = network_graph.channel(42).unwrap();
// Even after we tell the scorer we definitely have enough available liquidity, it will
// still remember that there was some failure in the past, and assign a non-0 penalty.
- scorer.payment_path_failed(&payment_path_for_amount(1000), 43);
+ scorer.payment_path_failed(&payment_path_for_amount(1000), 43, Duration::ZERO);
{
let network_graph = network_graph.read_only();
let channel = network_graph.channel(42).unwrap();
inflight_htlc_msat: 1024,
effective_capacity: EffectiveCapacity::Total { capacity_msat: 1_024, htlc_maximum_msat: 1_024 },
};
- scorer.payment_path_failed(&payment_path_for_amount(1), 42);
+ scorer.payment_path_failed(&payment_path_for_amount(1), 42, Duration::from_secs(10 * 16));
{
let network_graph = network_graph.read_only();
let channel = network_graph.channel(42).unwrap();
path_hop(source_pubkey(), 42, 1),
path_hop(sender_pubkey(), 41, 0),
];
- scorer.payment_path_failed(&Path { hops: path, blinded_tail: None }, 42);
+ scorer.payment_path_failed(&Path { hops: path, blinded_tail: None }, 42, Duration::from_secs(10 * (16 + 60 * 60)));
}
#[test]
// final value is taken into account.
assert!(scorer.channel_liquidities.get(&42).is_none());
- scorer.payment_path_failed(&path, 42);
+ scorer.payment_path_failed(&path, 42, Duration::ZERO);
path.blinded_tail.as_mut().unwrap().final_value_msat = 256;
- scorer.payment_path_failed(&path, 43);
+ scorer.payment_path_failed(&path, 43, Duration::ZERO);
let liquidity = scorer.channel_liquidities.get(&42).unwrap()
.as_directed(&source, &target, 1_000, decay_params);
None);
// Fail to pay once, and then check the buckets and penalty.
- scorer.payment_path_failed(&payment_path_for_amount(amount_msat), 42);
+ scorer.payment_path_failed(&payment_path_for_amount(amount_msat), 42, Duration::ZERO);
// The penalty should be the maximum penalty, as the payment we're scoring is now in the
// same bucket which is the only maximum datapoint.
assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms),
// ...but once we see a failure, we consider the payment to be substantially less likely,
// even though not a probability of zero as we still look at the second max bucket which
// now shows 31.
- scorer.payment_path_failed(&payment_path_for_amount(amount_msat), 42);
+ scorer.payment_path_failed(&payment_path_for_amount(amount_msat), 42, Duration::ZERO);
assert_eq!(scorer.historical_estimated_channel_liquidity_probabilities(42, &target),
Some(([63, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[32, 31, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0])));
projects / rust-lightning / blobdiff