mirror of
https://github.com/lightningdevkit/rust-lightning.git
synced 2025-03-10 21:36:17 +01:00
667 lines
30 KiB
Rust
667 lines
30 KiB
Rust
//! Utilities that take care of tasks that (1) need to happen periodically to keep Rust-Lightning
|
|
//! running properly, and (2) either can or should be run in the background. See docs for
|
|
//! [`BackgroundProcessor`] for more details on the nitty-gritty.
|
|
|
|
#![deny(broken_intra_doc_links)]
|
|
#![deny(missing_docs)]
|
|
#![deny(unsafe_code)]
|
|
|
|
#[macro_use] extern crate lightning;
|
|
|
|
use lightning::chain;
|
|
use lightning::chain::chaininterface::{BroadcasterInterface, FeeEstimator};
|
|
use lightning::chain::chainmonitor::{ChainMonitor, Persist};
|
|
use lightning::chain::keysinterface::{Sign, KeysInterface};
|
|
use lightning::ln::channelmanager::ChannelManager;
|
|
use lightning::ln::msgs::{ChannelMessageHandler, RoutingMessageHandler};
|
|
use lightning::ln::peer_handler::{CustomMessageHandler, PeerManager, SocketDescriptor};
|
|
use lightning::routing::network_graph::{NetworkGraph, NetGraphMsgHandler};
|
|
use lightning::util::events::{Event, EventHandler, EventsProvider};
|
|
use lightning::util::logger::Logger;
|
|
use std::sync::Arc;
|
|
use std::sync::atomic::{AtomicBool, Ordering};
|
|
use std::thread;
|
|
use std::thread::JoinHandle;
|
|
use std::time::{Duration, Instant};
|
|
use std::ops::Deref;
|
|
|
|
/// `BackgroundProcessor` takes care of tasks that (1) need to happen periodically to keep
|
|
/// Rust-Lightning running properly, and (2) either can or should be run in the background. Its
|
|
/// responsibilities are:
|
|
/// * Processing [`Event`]s with a user-provided [`EventHandler`].
|
|
/// * Monitoring whether the [`ChannelManager`] needs to be re-persisted to disk, and if so,
|
|
/// writing it to disk/backups by invoking the callback given to it at startup.
|
|
/// [`ChannelManager`] persistence should be done in the background.
|
|
/// * Calling [`ChannelManager::timer_tick_occurred`] and [`PeerManager::timer_tick_occurred`]
|
|
/// at the appropriate intervals.
|
|
/// * Calling [`NetworkGraph::remove_stale_channels`] (if a [`NetGraphMsgHandler`] is provided to
|
|
/// [`BackgroundProcessor::start`]).
|
|
///
|
|
/// It will also call [`PeerManager::process_events`] periodically though this shouldn't be relied
|
|
/// upon as doing so may result in high latency.
|
|
///
|
|
/// # Note
|
|
///
|
|
/// If [`ChannelManager`] persistence fails and the persisted manager becomes out-of-date, then
|
|
/// there is a risk of channels force-closing on startup when the manager realizes it's outdated.
|
|
/// However, as long as [`ChannelMonitor`] backups are sound, no funds besides those used for
|
|
/// unilateral chain closure fees are at risk.
|
|
///
|
|
/// [`ChannelMonitor`]: lightning::chain::channelmonitor::ChannelMonitor
|
|
/// [`Event`]: lightning::util::events::Event
|
|
#[must_use = "BackgroundProcessor will immediately stop on drop. It should be stored until shutdown."]
|
|
pub struct BackgroundProcessor {
|
|
stop_thread: Arc<AtomicBool>,
|
|
thread_handle: Option<JoinHandle<Result<(), std::io::Error>>>,
|
|
}
|
|
|
|
#[cfg(not(test))]
|
|
const FRESHNESS_TIMER: u64 = 60;
|
|
#[cfg(test)]
|
|
const FRESHNESS_TIMER: u64 = 1;
|
|
|
|
#[cfg(all(not(test), not(debug_assertions)))]
|
|
const PING_TIMER: u64 = 5;
|
|
/// Signature operations take a lot longer without compiler optimisations.
|
|
/// Increasing the ping timer allows for this but slower devices will be disconnected if the
|
|
/// timeout is reached.
|
|
#[cfg(all(not(test), debug_assertions))]
|
|
const PING_TIMER: u64 = 30;
|
|
#[cfg(test)]
|
|
const PING_TIMER: u64 = 1;
|
|
|
|
/// Prune the network graph of stale entries hourly.
|
|
const NETWORK_PRUNE_TIMER: u64 = 60 * 60;
|
|
|
|
/// Trait which handles persisting a [`ChannelManager`] to disk.
|
|
///
|
|
/// [`ChannelManager`]: lightning::ln::channelmanager::ChannelManager
|
|
pub trait ChannelManagerPersister<Signer: Sign, M: Deref, T: Deref, K: Deref, F: Deref, L: Deref>
|
|
where
|
|
M::Target: 'static + chain::Watch<Signer>,
|
|
T::Target: 'static + BroadcasterInterface,
|
|
K::Target: 'static + KeysInterface<Signer = Signer>,
|
|
F::Target: 'static + FeeEstimator,
|
|
L::Target: 'static + Logger,
|
|
{
|
|
/// Persist the given [`ChannelManager`] to disk, returning an error if persistence failed
|
|
/// (which will cause the [`BackgroundProcessor`] which called this method to exit.
|
|
///
|
|
/// [`ChannelManager`]: lightning::ln::channelmanager::ChannelManager
|
|
fn persist_manager(&self, channel_manager: &ChannelManager<Signer, M, T, K, F, L>) -> Result<(), std::io::Error>;
|
|
}
|
|
|
|
impl<Fun, Signer: Sign, M: Deref, T: Deref, K: Deref, F: Deref, L: Deref>
|
|
ChannelManagerPersister<Signer, M, T, K, F, L> for Fun where
|
|
M::Target: 'static + chain::Watch<Signer>,
|
|
T::Target: 'static + BroadcasterInterface,
|
|
K::Target: 'static + KeysInterface<Signer = Signer>,
|
|
F::Target: 'static + FeeEstimator,
|
|
L::Target: 'static + Logger,
|
|
Fun: Fn(&ChannelManager<Signer, M, T, K, F, L>) -> Result<(), std::io::Error>,
|
|
{
|
|
fn persist_manager(&self, channel_manager: &ChannelManager<Signer, M, T, K, F, L>) -> Result<(), std::io::Error> {
|
|
self(channel_manager)
|
|
}
|
|
}
|
|
|
|
/// Decorates an [`EventHandler`] with common functionality provided by standard [`EventHandler`]s.
|
|
struct DecoratingEventHandler<
|
|
E: EventHandler,
|
|
N: Deref<Target = NetGraphMsgHandler<G, A, L>>,
|
|
G: Deref<Target = NetworkGraph>,
|
|
A: Deref,
|
|
L: Deref,
|
|
>
|
|
where A::Target: chain::Access, L::Target: Logger {
|
|
event_handler: E,
|
|
net_graph_msg_handler: Option<N>,
|
|
}
|
|
|
|
impl<
|
|
E: EventHandler,
|
|
N: Deref<Target = NetGraphMsgHandler<G, A, L>>,
|
|
G: Deref<Target = NetworkGraph>,
|
|
A: Deref,
|
|
L: Deref,
|
|
> EventHandler for DecoratingEventHandler<E, N, G, A, L>
|
|
where A::Target: chain::Access, L::Target: Logger {
|
|
fn handle_event(&self, event: &Event) {
|
|
if let Some(event_handler) = &self.net_graph_msg_handler {
|
|
event_handler.handle_event(event);
|
|
}
|
|
self.event_handler.handle_event(event);
|
|
}
|
|
}
|
|
|
|
impl BackgroundProcessor {
|
|
/// Start a background thread that takes care of responsibilities enumerated in the [top-level
|
|
/// documentation].
|
|
///
|
|
/// The thread runs indefinitely unless the object is dropped, [`stop`] is called, or
|
|
/// `persist_manager` returns an error. In case of an error, the error is retrieved by calling
|
|
/// either [`join`] or [`stop`].
|
|
///
|
|
/// # Data Persistence
|
|
///
|
|
/// `persist_manager` is responsible for writing out the [`ChannelManager`] to disk, and/or
|
|
/// uploading to one or more backup services. See [`ChannelManager::write`] for writing out a
|
|
/// [`ChannelManager`]. See [`FilesystemPersister::persist_manager`] for Rust-Lightning's
|
|
/// provided implementation.
|
|
///
|
|
/// Typically, users should either implement [`ChannelManagerPersister`] to never return an
|
|
/// error or call [`join`] and handle any error that may arise. For the latter case,
|
|
/// `BackgroundProcessor` must be restarted by calling `start` again after handling the error.
|
|
///
|
|
/// # Event Handling
|
|
///
|
|
/// `event_handler` is responsible for handling events that users should be notified of (e.g.,
|
|
/// payment failed). [`BackgroundProcessor`] may decorate the given [`EventHandler`] with common
|
|
/// functionality implemented by other handlers.
|
|
/// * [`NetGraphMsgHandler`] if given will update the [`NetworkGraph`] based on payment failures.
|
|
///
|
|
/// [top-level documentation]: BackgroundProcessor
|
|
/// [`join`]: Self::join
|
|
/// [`stop`]: Self::stop
|
|
/// [`ChannelManager`]: lightning::ln::channelmanager::ChannelManager
|
|
/// [`ChannelManager::write`]: lightning::ln::channelmanager::ChannelManager#impl-Writeable
|
|
/// [`FilesystemPersister::persist_manager`]: lightning_persister::FilesystemPersister::persist_manager
|
|
/// [`NetworkGraph`]: lightning::routing::network_graph::NetworkGraph
|
|
pub fn start<
|
|
Signer: 'static + Sign,
|
|
CA: 'static + Deref + Send + Sync,
|
|
CF: 'static + Deref + Send + Sync,
|
|
CW: 'static + Deref + Send + Sync,
|
|
T: 'static + Deref + Send + Sync,
|
|
K: 'static + Deref + Send + Sync,
|
|
F: 'static + Deref + Send + Sync,
|
|
G: 'static + Deref<Target = NetworkGraph> + Send + Sync,
|
|
L: 'static + Deref + Send + Sync,
|
|
P: 'static + Deref + Send + Sync,
|
|
Descriptor: 'static + SocketDescriptor + Send + Sync,
|
|
CMH: 'static + Deref + Send + Sync,
|
|
RMH: 'static + Deref + Send + Sync,
|
|
EH: 'static + EventHandler + Send,
|
|
CMP: 'static + Send + ChannelManagerPersister<Signer, CW, T, K, F, L>,
|
|
M: 'static + Deref<Target = ChainMonitor<Signer, CF, T, F, L, P>> + Send + Sync,
|
|
CM: 'static + Deref<Target = ChannelManager<Signer, CW, T, K, F, L>> + Send + Sync,
|
|
NG: 'static + Deref<Target = NetGraphMsgHandler<G, CA, L>> + Send + Sync,
|
|
UMH: 'static + Deref + Send + Sync,
|
|
PM: 'static + Deref<Target = PeerManager<Descriptor, CMH, RMH, L, UMH>> + Send + Sync,
|
|
>(
|
|
persister: CMP, event_handler: EH, chain_monitor: M, channel_manager: CM,
|
|
net_graph_msg_handler: Option<NG>, peer_manager: PM, logger: L
|
|
) -> Self
|
|
where
|
|
CA::Target: 'static + chain::Access,
|
|
CF::Target: 'static + chain::Filter,
|
|
CW::Target: 'static + chain::Watch<Signer>,
|
|
T::Target: 'static + BroadcasterInterface,
|
|
K::Target: 'static + KeysInterface<Signer = Signer>,
|
|
F::Target: 'static + FeeEstimator,
|
|
L::Target: 'static + Logger,
|
|
P::Target: 'static + Persist<Signer>,
|
|
CMH::Target: 'static + ChannelMessageHandler,
|
|
RMH::Target: 'static + RoutingMessageHandler,
|
|
UMH::Target: 'static + CustomMessageHandler,
|
|
{
|
|
let stop_thread = Arc::new(AtomicBool::new(false));
|
|
let stop_thread_clone = stop_thread.clone();
|
|
let handle = thread::spawn(move || -> Result<(), std::io::Error> {
|
|
let event_handler = DecoratingEventHandler { event_handler, net_graph_msg_handler: net_graph_msg_handler.as_ref().map(|t| t.deref()) };
|
|
|
|
log_trace!(logger, "Calling ChannelManager's timer_tick_occurred on startup");
|
|
channel_manager.timer_tick_occurred();
|
|
|
|
let mut last_freshness_call = Instant::now();
|
|
let mut last_ping_call = Instant::now();
|
|
let mut last_prune_call = Instant::now();
|
|
let mut have_pruned = false;
|
|
|
|
loop {
|
|
peer_manager.process_events();
|
|
channel_manager.process_pending_events(&event_handler);
|
|
chain_monitor.process_pending_events(&event_handler);
|
|
let updates_available =
|
|
channel_manager.await_persistable_update_timeout(Duration::from_millis(100));
|
|
if updates_available {
|
|
log_trace!(logger, "Persisting ChannelManager...");
|
|
persister.persist_manager(&*channel_manager)?;
|
|
log_trace!(logger, "Done persisting ChannelManager.");
|
|
}
|
|
// Exit the loop if the background processor was requested to stop.
|
|
if stop_thread.load(Ordering::Acquire) == true {
|
|
log_trace!(logger, "Terminating background processor.");
|
|
return Ok(());
|
|
}
|
|
if last_freshness_call.elapsed().as_secs() > FRESHNESS_TIMER {
|
|
log_trace!(logger, "Calling ChannelManager's timer_tick_occurred");
|
|
channel_manager.timer_tick_occurred();
|
|
last_freshness_call = Instant::now();
|
|
}
|
|
if last_ping_call.elapsed().as_secs() > PING_TIMER * 2 {
|
|
// On various platforms, we may be starved of CPU cycles for several reasons.
|
|
// E.g. on iOS, if we've been in the background, we will be entirely paused.
|
|
// Similarly, if we're on a desktop platform and the device has been asleep, we
|
|
// may not get any cycles.
|
|
// In any case, if we've been entirely paused for more than double our ping
|
|
// timer, we should have disconnected all sockets by now (and they're probably
|
|
// dead anyway), so disconnect them by calling `timer_tick_occurred()` twice.
|
|
log_trace!(logger, "Awoke after more than double our ping timer, disconnecting peers.");
|
|
peer_manager.disconnect_all_peers();
|
|
last_ping_call = Instant::now();
|
|
} else if last_ping_call.elapsed().as_secs() > PING_TIMER {
|
|
log_trace!(logger, "Calling PeerManager's timer_tick_occurred");
|
|
peer_manager.timer_tick_occurred();
|
|
last_ping_call = Instant::now();
|
|
}
|
|
|
|
// Note that we want to run a graph prune once not long after startup before
|
|
// falling back to our usual hourly prunes. This avoids short-lived clients never
|
|
// pruning their network graph. We run once 60 seconds after startup before
|
|
// continuing our normal cadence.
|
|
if last_prune_call.elapsed().as_secs() > if have_pruned { NETWORK_PRUNE_TIMER } else { 60 } {
|
|
if let Some(ref handler) = net_graph_msg_handler {
|
|
log_trace!(logger, "Pruning network graph of stale entries");
|
|
handler.network_graph().remove_stale_channels();
|
|
last_prune_call = Instant::now();
|
|
have_pruned = true;
|
|
}
|
|
}
|
|
}
|
|
});
|
|
Self { stop_thread: stop_thread_clone, thread_handle: Some(handle) }
|
|
}
|
|
|
|
/// Join `BackgroundProcessor`'s thread, returning any error that occurred while persisting
|
|
/// [`ChannelManager`].
|
|
///
|
|
/// # Panics
|
|
///
|
|
/// This function panics if the background thread has panicked such as while persisting or
|
|
/// handling events.
|
|
///
|
|
/// [`ChannelManager`]: lightning::ln::channelmanager::ChannelManager
|
|
pub fn join(mut self) -> Result<(), std::io::Error> {
|
|
assert!(self.thread_handle.is_some());
|
|
self.join_thread()
|
|
}
|
|
|
|
/// Stop `BackgroundProcessor`'s thread, returning any error that occurred while persisting
|
|
/// [`ChannelManager`].
|
|
///
|
|
/// # Panics
|
|
///
|
|
/// This function panics if the background thread has panicked such as while persisting or
|
|
/// handling events.
|
|
///
|
|
/// [`ChannelManager`]: lightning::ln::channelmanager::ChannelManager
|
|
pub fn stop(mut self) -> Result<(), std::io::Error> {
|
|
assert!(self.thread_handle.is_some());
|
|
self.stop_and_join_thread()
|
|
}
|
|
|
|
fn stop_and_join_thread(&mut self) -> Result<(), std::io::Error> {
|
|
self.stop_thread.store(true, Ordering::Release);
|
|
self.join_thread()
|
|
}
|
|
|
|
fn join_thread(&mut self) -> Result<(), std::io::Error> {
|
|
match self.thread_handle.take() {
|
|
Some(handle) => handle.join().unwrap(),
|
|
None => Ok(()),
|
|
}
|
|
}
|
|
}
|
|
|
|
impl Drop for BackgroundProcessor {
|
|
fn drop(&mut self) {
|
|
self.stop_and_join_thread().unwrap();
|
|
}
|
|
}
|
|
|
|
#[cfg(test)]
|
|
mod tests {
|
|
use bitcoin::blockdata::block::BlockHeader;
|
|
use bitcoin::blockdata::constants::genesis_block;
|
|
use bitcoin::blockdata::transaction::{Transaction, TxOut};
|
|
use bitcoin::network::constants::Network;
|
|
use lightning::chain::{BestBlock, Confirm, chainmonitor};
|
|
use lightning::chain::channelmonitor::ANTI_REORG_DELAY;
|
|
use lightning::chain::keysinterface::{InMemorySigner, KeysInterface, KeysManager};
|
|
use lightning::chain::transaction::OutPoint;
|
|
use lightning::get_event_msg;
|
|
use lightning::ln::channelmanager::{BREAKDOWN_TIMEOUT, ChainParameters, ChannelManager, SimpleArcChannelManager};
|
|
use lightning::ln::features::InitFeatures;
|
|
use lightning::ln::msgs::{ChannelMessageHandler, Init};
|
|
use lightning::ln::peer_handler::{PeerManager, MessageHandler, SocketDescriptor, IgnoringMessageHandler};
|
|
use lightning::routing::network_graph::{NetworkGraph, NetGraphMsgHandler};
|
|
use lightning::util::config::UserConfig;
|
|
use lightning::util::events::{Event, MessageSendEventsProvider, MessageSendEvent};
|
|
use lightning::util::ser::Writeable;
|
|
use lightning::util::test_utils;
|
|
use lightning_invoice::payment::{InvoicePayer, RetryAttempts};
|
|
use lightning_invoice::utils::DefaultRouter;
|
|
use lightning_persister::FilesystemPersister;
|
|
use std::fs;
|
|
use std::path::PathBuf;
|
|
use std::sync::{Arc, Mutex};
|
|
use std::time::Duration;
|
|
use super::{BackgroundProcessor, FRESHNESS_TIMER};
|
|
|
|
const EVENT_DEADLINE: u64 = 5 * FRESHNESS_TIMER;
|
|
|
|
#[derive(Clone, Eq, Hash, PartialEq)]
|
|
struct TestDescriptor{}
|
|
impl SocketDescriptor for TestDescriptor {
|
|
fn send_data(&mut self, _data: &[u8], _resume_read: bool) -> usize {
|
|
0
|
|
}
|
|
|
|
fn disconnect_socket(&mut self) {}
|
|
}
|
|
|
|
type ChainMonitor = chainmonitor::ChainMonitor<InMemorySigner, Arc<test_utils::TestChainSource>, Arc<test_utils::TestBroadcaster>, Arc<test_utils::TestFeeEstimator>, Arc<test_utils::TestLogger>, Arc<FilesystemPersister>>;
|
|
|
|
struct Node {
|
|
node: Arc<SimpleArcChannelManager<ChainMonitor, test_utils::TestBroadcaster, test_utils::TestFeeEstimator, test_utils::TestLogger>>,
|
|
net_graph_msg_handler: Option<Arc<NetGraphMsgHandler<Arc<NetworkGraph>, Arc<test_utils::TestChainSource>, Arc<test_utils::TestLogger>>>>,
|
|
peer_manager: Arc<PeerManager<TestDescriptor, Arc<test_utils::TestChannelMessageHandler>, Arc<test_utils::TestRoutingMessageHandler>, Arc<test_utils::TestLogger>, IgnoringMessageHandler>>,
|
|
chain_monitor: Arc<ChainMonitor>,
|
|
persister: Arc<FilesystemPersister>,
|
|
tx_broadcaster: Arc<test_utils::TestBroadcaster>,
|
|
network_graph: Arc<NetworkGraph>,
|
|
logger: Arc<test_utils::TestLogger>,
|
|
best_block: BestBlock,
|
|
}
|
|
|
|
impl Drop for Node {
|
|
fn drop(&mut self) {
|
|
let data_dir = self.persister.get_data_dir();
|
|
match fs::remove_dir_all(data_dir.clone()) {
|
|
Err(e) => println!("Failed to remove test persister directory {}: {}", data_dir, e),
|
|
_ => {}
|
|
}
|
|
}
|
|
}
|
|
|
|
fn get_full_filepath(filepath: String, filename: String) -> String {
|
|
let mut path = PathBuf::from(filepath);
|
|
path.push(filename);
|
|
path.to_str().unwrap().to_string()
|
|
}
|
|
|
|
fn create_nodes(num_nodes: usize, persist_dir: String) -> Vec<Node> {
|
|
let mut nodes = Vec::new();
|
|
for i in 0..num_nodes {
|
|
let tx_broadcaster = Arc::new(test_utils::TestBroadcaster{txn_broadcasted: Mutex::new(Vec::new()), blocks: Arc::new(Mutex::new(Vec::new()))});
|
|
let fee_estimator = Arc::new(test_utils::TestFeeEstimator { sat_per_kw: Mutex::new(253) });
|
|
let chain_source = Arc::new(test_utils::TestChainSource::new(Network::Testnet));
|
|
let logger = Arc::new(test_utils::TestLogger::with_id(format!("node {}", i)));
|
|
let persister = Arc::new(FilesystemPersister::new(format!("{}_persister_{}", persist_dir, i)));
|
|
let seed = [i as u8; 32];
|
|
let network = Network::Testnet;
|
|
let genesis_block = genesis_block(network);
|
|
let now = Duration::from_secs(genesis_block.header.time as u64);
|
|
let keys_manager = Arc::new(KeysManager::new(&seed, now.as_secs(), now.subsec_nanos()));
|
|
let chain_monitor = Arc::new(chainmonitor::ChainMonitor::new(Some(chain_source.clone()), tx_broadcaster.clone(), logger.clone(), fee_estimator.clone(), persister.clone()));
|
|
let best_block = BestBlock::from_genesis(network);
|
|
let params = ChainParameters { network, best_block };
|
|
let manager = Arc::new(ChannelManager::new(fee_estimator.clone(), chain_monitor.clone(), tx_broadcaster.clone(), logger.clone(), keys_manager.clone(), UserConfig::default(), params));
|
|
let network_graph = Arc::new(NetworkGraph::new(genesis_block.header.block_hash()));
|
|
let net_graph_msg_handler = Some(Arc::new(NetGraphMsgHandler::new(network_graph.clone(), Some(chain_source.clone()), logger.clone())));
|
|
let msg_handler = MessageHandler { chan_handler: Arc::new(test_utils::TestChannelMessageHandler::new()), route_handler: Arc::new(test_utils::TestRoutingMessageHandler::new() )};
|
|
let peer_manager = Arc::new(PeerManager::new(msg_handler, keys_manager.get_node_secret(), &seed, logger.clone(), IgnoringMessageHandler{}));
|
|
let node = Node { node: manager, net_graph_msg_handler, peer_manager, chain_monitor, persister, tx_broadcaster, network_graph, logger, best_block };
|
|
nodes.push(node);
|
|
}
|
|
|
|
for i in 0..num_nodes {
|
|
for j in (i+1)..num_nodes {
|
|
nodes[i].node.peer_connected(&nodes[j].node.get_our_node_id(), &Init { features: InitFeatures::known() });
|
|
nodes[j].node.peer_connected(&nodes[i].node.get_our_node_id(), &Init { features: InitFeatures::known() });
|
|
}
|
|
}
|
|
|
|
nodes
|
|
}
|
|
|
|
macro_rules! open_channel {
|
|
($node_a: expr, $node_b: expr, $channel_value: expr) => {{
|
|
begin_open_channel!($node_a, $node_b, $channel_value);
|
|
let events = $node_a.node.get_and_clear_pending_events();
|
|
assert_eq!(events.len(), 1);
|
|
let (temporary_channel_id, tx) = handle_funding_generation_ready!(&events[0], $channel_value);
|
|
end_open_channel!($node_a, $node_b, temporary_channel_id, tx);
|
|
tx
|
|
}}
|
|
}
|
|
|
|
macro_rules! begin_open_channel {
|
|
($node_a: expr, $node_b: expr, $channel_value: expr) => {{
|
|
$node_a.node.create_channel($node_b.node.get_our_node_id(), $channel_value, 100, 42, None).unwrap();
|
|
$node_b.node.handle_open_channel(&$node_a.node.get_our_node_id(), InitFeatures::known(), &get_event_msg!($node_a, MessageSendEvent::SendOpenChannel, $node_b.node.get_our_node_id()));
|
|
$node_a.node.handle_accept_channel(&$node_b.node.get_our_node_id(), InitFeatures::known(), &get_event_msg!($node_b, MessageSendEvent::SendAcceptChannel, $node_a.node.get_our_node_id()));
|
|
}}
|
|
}
|
|
|
|
macro_rules! handle_funding_generation_ready {
|
|
($event: expr, $channel_value: expr) => {{
|
|
match $event {
|
|
&Event::FundingGenerationReady { temporary_channel_id, channel_value_satoshis, ref output_script, user_channel_id } => {
|
|
assert_eq!(channel_value_satoshis, $channel_value);
|
|
assert_eq!(user_channel_id, 42);
|
|
|
|
let tx = Transaction { version: 1 as i32, lock_time: 0, input: Vec::new(), output: vec![TxOut {
|
|
value: channel_value_satoshis, script_pubkey: output_script.clone(),
|
|
}]};
|
|
(temporary_channel_id, tx)
|
|
},
|
|
_ => panic!("Unexpected event"),
|
|
}
|
|
}}
|
|
}
|
|
|
|
macro_rules! end_open_channel {
|
|
($node_a: expr, $node_b: expr, $temporary_channel_id: expr, $tx: expr) => {{
|
|
$node_a.node.funding_transaction_generated(&$temporary_channel_id, $tx.clone()).unwrap();
|
|
$node_b.node.handle_funding_created(&$node_a.node.get_our_node_id(), &get_event_msg!($node_a, MessageSendEvent::SendFundingCreated, $node_b.node.get_our_node_id()));
|
|
$node_a.node.handle_funding_signed(&$node_b.node.get_our_node_id(), &get_event_msg!($node_b, MessageSendEvent::SendFundingSigned, $node_a.node.get_our_node_id()));
|
|
}}
|
|
}
|
|
|
|
fn confirm_transaction_depth(node: &mut Node, tx: &Transaction, depth: u32) {
|
|
for i in 1..=depth {
|
|
let prev_blockhash = node.best_block.block_hash();
|
|
let height = node.best_block.height() + 1;
|
|
let header = BlockHeader { version: 0x20000000, prev_blockhash, merkle_root: Default::default(), time: height, bits: 42, nonce: 42 };
|
|
let txdata = vec![(0, tx)];
|
|
node.best_block = BestBlock::new(header.block_hash(), height);
|
|
match i {
|
|
1 => {
|
|
node.node.transactions_confirmed(&header, &txdata, height);
|
|
node.chain_monitor.transactions_confirmed(&header, &txdata, height);
|
|
},
|
|
x if x == depth => {
|
|
node.node.best_block_updated(&header, height);
|
|
node.chain_monitor.best_block_updated(&header, height);
|
|
},
|
|
_ => {},
|
|
}
|
|
}
|
|
}
|
|
fn confirm_transaction(node: &mut Node, tx: &Transaction) {
|
|
confirm_transaction_depth(node, tx, ANTI_REORG_DELAY);
|
|
}
|
|
|
|
#[test]
|
|
fn test_background_processor() {
|
|
// Test that when a new channel is created, the ChannelManager needs to be re-persisted with
|
|
// updates. Also test that when new updates are available, the manager signals that it needs
|
|
// re-persistence and is successfully re-persisted.
|
|
let nodes = create_nodes(2, "test_background_processor".to_string());
|
|
|
|
// Go through the channel creation process so that each node has something to persist. Since
|
|
// open_channel consumes events, it must complete before starting BackgroundProcessor to
|
|
// avoid a race with processing events.
|
|
let tx = open_channel!(nodes[0], nodes[1], 100000);
|
|
|
|
// Initiate the background processors to watch each node.
|
|
let data_dir = nodes[0].persister.get_data_dir();
|
|
let persister = move |node: &ChannelManager<InMemorySigner, Arc<ChainMonitor>, Arc<test_utils::TestBroadcaster>, Arc<KeysManager>, Arc<test_utils::TestFeeEstimator>, Arc<test_utils::TestLogger>>| FilesystemPersister::persist_manager(data_dir.clone(), node);
|
|
let event_handler = |_: &_| {};
|
|
let bg_processor = BackgroundProcessor::start(persister, event_handler, nodes[0].chain_monitor.clone(), nodes[0].node.clone(), nodes[0].net_graph_msg_handler.clone(), nodes[0].peer_manager.clone(), nodes[0].logger.clone());
|
|
|
|
macro_rules! check_persisted_data {
|
|
($node: expr, $filepath: expr, $expected_bytes: expr) => {
|
|
loop {
|
|
$expected_bytes.clear();
|
|
match $node.write(&mut $expected_bytes) {
|
|
Ok(()) => {
|
|
match std::fs::read($filepath) {
|
|
Ok(bytes) => {
|
|
if bytes == $expected_bytes {
|
|
break
|
|
} else {
|
|
continue
|
|
}
|
|
},
|
|
Err(_) => continue
|
|
}
|
|
},
|
|
Err(e) => panic!("Unexpected error: {}", e)
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
// Check that the initial channel manager data is persisted as expected.
|
|
let filepath = get_full_filepath("test_background_processor_persister_0".to_string(), "manager".to_string());
|
|
let mut expected_bytes = Vec::new();
|
|
check_persisted_data!(nodes[0].node, filepath.clone(), expected_bytes);
|
|
loop {
|
|
if !nodes[0].node.get_persistence_condvar_value() { break }
|
|
}
|
|
|
|
// Force-close the channel.
|
|
nodes[0].node.force_close_channel(&OutPoint { txid: tx.txid(), index: 0 }.to_channel_id()).unwrap();
|
|
|
|
// Check that the force-close updates are persisted.
|
|
let mut expected_bytes = Vec::new();
|
|
check_persisted_data!(nodes[0].node, filepath.clone(), expected_bytes);
|
|
loop {
|
|
if !nodes[0].node.get_persistence_condvar_value() { break }
|
|
}
|
|
|
|
assert!(bg_processor.stop().is_ok());
|
|
}
|
|
|
|
#[test]
|
|
fn test_timer_tick_called() {
|
|
// Test that ChannelManager's and PeerManager's `timer_tick_occurred` is called every
|
|
// `FRESHNESS_TIMER`.
|
|
let nodes = create_nodes(1, "test_timer_tick_called".to_string());
|
|
let data_dir = nodes[0].persister.get_data_dir();
|
|
let persister = move |node: &ChannelManager<InMemorySigner, Arc<ChainMonitor>, Arc<test_utils::TestBroadcaster>, Arc<KeysManager>, Arc<test_utils::TestFeeEstimator>, Arc<test_utils::TestLogger>>| FilesystemPersister::persist_manager(data_dir.clone(), node);
|
|
let event_handler = |_: &_| {};
|
|
let bg_processor = BackgroundProcessor::start(persister, event_handler, nodes[0].chain_monitor.clone(), nodes[0].node.clone(), nodes[0].net_graph_msg_handler.clone(), nodes[0].peer_manager.clone(), nodes[0].logger.clone());
|
|
loop {
|
|
let log_entries = nodes[0].logger.lines.lock().unwrap();
|
|
let desired_log = "Calling ChannelManager's timer_tick_occurred".to_string();
|
|
let second_desired_log = "Calling PeerManager's timer_tick_occurred".to_string();
|
|
if log_entries.get(&("lightning_background_processor".to_string(), desired_log)).is_some() &&
|
|
log_entries.get(&("lightning_background_processor".to_string(), second_desired_log)).is_some() {
|
|
break
|
|
}
|
|
}
|
|
|
|
assert!(bg_processor.stop().is_ok());
|
|
}
|
|
|
|
#[test]
|
|
fn test_persist_error() {
|
|
// Test that if we encounter an error during manager persistence, the thread panics.
|
|
let nodes = create_nodes(2, "test_persist_error".to_string());
|
|
open_channel!(nodes[0], nodes[1], 100000);
|
|
|
|
let persister = |_: &_| Err(std::io::Error::new(std::io::ErrorKind::Other, "test"));
|
|
let event_handler = |_: &_| {};
|
|
let bg_processor = BackgroundProcessor::start(persister, event_handler, nodes[0].chain_monitor.clone(), nodes[0].node.clone(), nodes[0].net_graph_msg_handler.clone(), nodes[0].peer_manager.clone(), nodes[0].logger.clone());
|
|
match bg_processor.join() {
|
|
Ok(_) => panic!("Expected error persisting manager"),
|
|
Err(e) => {
|
|
assert_eq!(e.kind(), std::io::ErrorKind::Other);
|
|
assert_eq!(e.get_ref().unwrap().to_string(), "test");
|
|
},
|
|
}
|
|
}
|
|
|
|
#[test]
|
|
fn test_background_event_handling() {
|
|
let mut nodes = create_nodes(2, "test_background_event_handling".to_string());
|
|
let channel_value = 100000;
|
|
let data_dir = nodes[0].persister.get_data_dir();
|
|
let persister = move |node: &_| FilesystemPersister::persist_manager(data_dir.clone(), node);
|
|
|
|
// Set up a background event handler for FundingGenerationReady events.
|
|
let (sender, receiver) = std::sync::mpsc::sync_channel(1);
|
|
let event_handler = move |event: &Event| {
|
|
sender.send(handle_funding_generation_ready!(event, channel_value)).unwrap();
|
|
};
|
|
let bg_processor = BackgroundProcessor::start(persister.clone(), event_handler, nodes[0].chain_monitor.clone(), nodes[0].node.clone(), nodes[0].net_graph_msg_handler.clone(), nodes[0].peer_manager.clone(), nodes[0].logger.clone());
|
|
|
|
// Open a channel and check that the FundingGenerationReady event was handled.
|
|
begin_open_channel!(nodes[0], nodes[1], channel_value);
|
|
let (temporary_channel_id, funding_tx) = receiver
|
|
.recv_timeout(Duration::from_secs(EVENT_DEADLINE))
|
|
.expect("FundingGenerationReady not handled within deadline");
|
|
end_open_channel!(nodes[0], nodes[1], temporary_channel_id, funding_tx);
|
|
|
|
// Confirm the funding transaction.
|
|
confirm_transaction(&mut nodes[0], &funding_tx);
|
|
let as_funding = get_event_msg!(nodes[0], MessageSendEvent::SendFundingLocked, nodes[1].node.get_our_node_id());
|
|
confirm_transaction(&mut nodes[1], &funding_tx);
|
|
let bs_funding = get_event_msg!(nodes[1], MessageSendEvent::SendFundingLocked, nodes[0].node.get_our_node_id());
|
|
nodes[0].node.handle_funding_locked(&nodes[1].node.get_our_node_id(), &bs_funding);
|
|
let _as_channel_update = get_event_msg!(nodes[0], MessageSendEvent::SendChannelUpdate, nodes[1].node.get_our_node_id());
|
|
nodes[1].node.handle_funding_locked(&nodes[0].node.get_our_node_id(), &as_funding);
|
|
let _bs_channel_update = get_event_msg!(nodes[1], MessageSendEvent::SendChannelUpdate, nodes[0].node.get_our_node_id());
|
|
|
|
assert!(bg_processor.stop().is_ok());
|
|
|
|
// Set up a background event handler for SpendableOutputs events.
|
|
let (sender, receiver) = std::sync::mpsc::sync_channel(1);
|
|
let event_handler = move |event: &Event| sender.send(event.clone()).unwrap();
|
|
let bg_processor = BackgroundProcessor::start(persister, event_handler, nodes[0].chain_monitor.clone(), nodes[0].node.clone(), nodes[0].net_graph_msg_handler.clone(), nodes[0].peer_manager.clone(), nodes[0].logger.clone());
|
|
|
|
// Force close the channel and check that the SpendableOutputs event was handled.
|
|
nodes[0].node.force_close_channel(&nodes[0].node.list_channels()[0].channel_id).unwrap();
|
|
let commitment_tx = nodes[0].tx_broadcaster.txn_broadcasted.lock().unwrap().pop().unwrap();
|
|
confirm_transaction_depth(&mut nodes[0], &commitment_tx, BREAKDOWN_TIMEOUT as u32);
|
|
let event = receiver
|
|
.recv_timeout(Duration::from_secs(EVENT_DEADLINE))
|
|
.expect("SpendableOutputs not handled within deadline");
|
|
match event {
|
|
Event::SpendableOutputs { .. } => {},
|
|
Event::ChannelClosed { .. } => {},
|
|
_ => panic!("Unexpected event: {:?}", event),
|
|
}
|
|
|
|
assert!(bg_processor.stop().is_ok());
|
|
}
|
|
|
|
#[test]
|
|
fn test_invoice_payer() {
|
|
let nodes = create_nodes(2, "test_invoice_payer".to_string());
|
|
|
|
// Initiate the background processors to watch each node.
|
|
let data_dir = nodes[0].persister.get_data_dir();
|
|
let persister = move |node: &ChannelManager<InMemorySigner, Arc<ChainMonitor>, Arc<test_utils::TestBroadcaster>, Arc<KeysManager>, Arc<test_utils::TestFeeEstimator>, Arc<test_utils::TestLogger>>| FilesystemPersister::persist_manager(data_dir.clone(), node);
|
|
let router = DefaultRouter::new(Arc::clone(&nodes[0].network_graph), Arc::clone(&nodes[0].logger));
|
|
let scorer = Arc::new(Mutex::new(test_utils::TestScorer::default()));
|
|
let invoice_payer = Arc::new(InvoicePayer::new(Arc::clone(&nodes[0].node), router, scorer, Arc::clone(&nodes[0].logger), |_: &_| {}, RetryAttempts(2)));
|
|
let event_handler = Arc::clone(&invoice_payer);
|
|
let bg_processor = BackgroundProcessor::start(persister, event_handler, nodes[0].chain_monitor.clone(), nodes[0].node.clone(), nodes[0].net_graph_msg_handler.clone(), nodes[0].peer_manager.clone(), nodes[0].logger.clone());
|
|
assert!(bg_processor.stop().is_ok());
|
|
}
|
|
}
|