Implement graceful shutdown (#2812)

* Remove stop status mutex
* remove some deadlocks
* Rewrite stop channel handling
* fix deadlock in peers object
* add missing test fixes

api/src/types.rs

@@ -434,8 +434,7 @@ impl<'de> serde::de::Deserialize<'de> for OutputPrintable {
 				}
 
 				if output_type.is_none()
-					|| commit.is_none()
+					|| commit.is_none() || spent.is_none()
-					|| spent.is_none()
 					|| proof_hash.is_none()
 					|| mmr_index.is_none()
 				{

chain/src/chain.rs

@@ -33,7 +33,7 @@ use crate::types::{
 	BlockStatus, ChainAdapter, NoStatus, Options, Tip, TxHashSetRoots, TxHashsetWriteStatus,
 };
 use crate::util::secp::pedersen::{Commitment, RangeProof};
-use crate::util::{Mutex, RwLock, StopState};
+use crate::util::RwLock;
 use grin_store::Error::NotFoundErr;
 use std::collections::HashMap;
 use std::fs::{self, File};
@@ -152,7 +152,6 @@ pub struct Chain {
 	// POW verification function
 	pow_verifier: fn(&BlockHeader) -> Result<(), pow::Error>,
 	archive_mode: bool,
-	stop_state: Arc<Mutex<StopState>>,
 	genesis: BlockHeader,
 }
 
@@ -167,16 +166,7 @@ impl Chain {
 		pow_verifier: fn(&BlockHeader) -> Result<(), pow::Error>,
 		verifier_cache: Arc<RwLock<dyn VerifierCache>>,
 		archive_mode: bool,
-		stop_state: Arc<Mutex<StopState>>,
 	) -> Result<Chain, Error> {
-		// Note: We take a lock on the stop_state here and do not release it until
-		// we have finished chain initialization.
-		let stop_state_local = stop_state.clone();
-		let stop_lock = stop_state_local.lock();
-		if stop_lock.is_stopped() {
-			return Err(ErrorKind::Stopped.into());
-		}
-
 		let store = Arc::new(store::ChainStore::new(&db_root)?);
 
 		// open the txhashset, creating a new one if necessary
@@ -194,7 +184,6 @@ impl Chain {
 			pow_verifier,
 			verifier_cache,
 			archive_mode,
-			stop_state,
 			genesis: genesis.header.clone(),
 		})
 	}
@@ -283,15 +272,6 @@ impl Chain {
 	/// or false if it has added to a fork (or orphan?).
 	fn process_block_single(&self, b: Block, opts: Options) -> Result<Option<Tip>, Error> {
 		let (maybe_new_head, prev_head) = {
-			// Note: We take a lock on the stop_state here and do not release it until
-			// we have finished processing this single block.
-			// We take care to write both the txhashset *and* the batch while we
-			// have the stop_state lock.
-			let stop_lock = self.stop_state.lock();
-			if stop_lock.is_stopped() {
-				return Err(ErrorKind::Stopped.into());
-			}
-
 			let mut txhashset = self.txhashset.write();
 			let batch = self.store.batch()?;
 			let mut ctx = self.new_ctx(opts, batch, &mut txhashset)?;
@@ -381,15 +361,6 @@ impl Chain {
 	/// This is only ever used during sync and is based on sync_head.
 	/// We update header_head here if our total work increases.
 	pub fn sync_block_headers(&self, headers: &[BlockHeader], opts: Options) -> Result<(), Error> {
-		// Note: We take a lock on the stop_state here and do not release it until
-		// we have finished processing this single block.
-		// We take care to write both the txhashset *and* the batch while we
-		// have the stop_state lock.
-		let stop_lock = self.stop_state.lock();
-		if stop_lock.is_stopped() {
-			return Err(ErrorKind::Stopped.into());
-		}
-
 		let mut txhashset = self.txhashset.write();
 		let batch = self.store.batch()?;
 		let mut ctx = self.new_ctx(opts, batch, &mut txhashset)?;
@@ -1098,14 +1069,6 @@ impl Chain {
 			}
 		}
 
-		// Note: We take a lock on the stop_state here and do not release it until
-		// we have finished processing this chain compaction operation.
-		// We want to avoid shutting the node down in the middle of compacting the data.
-		let stop_lock = self.stop_state.lock();
-		if stop_lock.is_stopped() {
-			return Err(ErrorKind::Stopped.into());
-		}
-
 		// Take a write lock on the txhashet and start a new writeable db batch.
 		let mut txhashset = self.txhashset.write();
 		let mut batch = self.store.batch()?;

chain/src/store.rs

@@ -384,7 +384,6 @@ impl<'a> Batch<'a> {
 		let key = to_key(BLOCK_PREFIX, &mut "".to_string().into_bytes());
 		self.db.iter(&key)
 	}
-	
 }
 
 /// An iterator on blocks, from latest to earliest, specialized to return

chain/tests/data_file_integrity.rs

@@ -21,7 +21,7 @@ use self::core::libtx;
 use self::core::pow::{self, Difficulty};
 use self::core::{consensus, genesis};
 use self::keychain::{ExtKeychain, ExtKeychainPath, Keychain};
-use self::util::{Mutex, RwLock, StopState};
+use self::util::RwLock;
 use chrono::Duration;
 use grin_chain as chain;
 use grin_core as core;
@@ -47,7 +47,6 @@ fn setup(dir_name: &str) -> Chain {
 		pow::verify_size,
 		verifier_cache,
 		false,
-		Arc::new(Mutex::new(StopState::new())),
 	)
 	.unwrap()
 }
@@ -61,7 +60,6 @@ fn reload_chain(dir_name: &str) -> Chain {
 		pow::verify_size,
 		verifier_cache,
 		false,
-		Arc::new(Mutex::new(StopState::new())),
 	)
 	.unwrap()
 }

chain/tests/mine_simple_chain.rs

@@ -23,7 +23,7 @@ use self::core::libtx::{self, build, reward};
 use self::core::pow::Difficulty;
 use self::core::{consensus, global, pow};
 use self::keychain::{ExtKeychain, ExtKeychainPath, Keychain};
-use self::util::{Mutex, RwLock, StopState};
+use self::util::{RwLock, StopState};
 use chrono::Duration;
 use grin_chain as chain;
 use grin_core as core;
@@ -47,7 +47,6 @@ fn setup(dir_name: &str, genesis: Block) -> Chain {
 		pow::verify_size,
 		verifier_cache,
 		false,
-		Arc::new(Mutex::new(StopState::new())),
 	)
 	.unwrap()
 }
@@ -565,7 +564,6 @@ fn actual_diff_iter_output() {
 		pow::verify_size,
 		verifier_cache,
 		false,
-		Arc::new(Mutex::new(StopState::new())),
 	)
 	.unwrap();
 	let iter = chain.difficulty_iter().unwrap();

chain/tests/test_coinbase_maturity.rs

@@ -20,7 +20,7 @@ use self::core::libtx::{self, build};
 use self::core::pow::Difficulty;
 use self::core::{consensus, pow};
 use self::keychain::{ExtKeychain, ExtKeychainPath, Keychain};
-use self::util::{Mutex, RwLock, StopState};
+use self::util::{RwLock, StopState};
 use chrono::Duration;
 use env_logger;
 use grin_chain as chain;
@@ -53,7 +53,6 @@ fn test_coinbase_maturity() {
 			pow::verify_size,
 			verifier_cache,
 			false,
-			Arc::new(Mutex::new(StopState::new())),
 		)
 		.unwrap();
 

etc/gen_gen/src/bin/gen_gen.rs

@@ -281,7 +281,7 @@ fn setup_chain(dir_name: &str, genesis: core::core::Block) -> chain::Chain {
 		core::pow::verify_size,
 		verifier_cache,
 		false,
-		Arc::new(util::Mutex::new(util::StopState::new())),
+		Arc::new(util::StopState::new()),
 	)
 	.unwrap()
 }

p2p/src/conn.rs

@@ -24,7 +24,11 @@ use std::fs::File;
 use std::io::{self, Read, Write};
 use std::net::{Shutdown, TcpStream};
 use std::sync::{mpsc, Arc};
-use std::{cmp, thread, time};
+use std::{
+	cmp,
+	thread::{self, JoinHandle},
+	time,
+};
 
 use crate::core::ser;
 use crate::core::ser::FixedLength;
@@ -43,7 +47,7 @@ pub trait MessageHandler: Send + 'static {
 		&self,
 		msg: Message<'a>,
 		writer: &'a mut dyn Write,
-		received_bytes: Arc<RwLock<RateCounter>>,
+		tracker: Arc<Tracker>,
 	) -> Result<Option<Response<'a>>, Error>;
 }
 
@@ -130,15 +134,12 @@ impl<'a> Response<'a> {
 		})
 	}
 
-	fn write(mut self, sent_bytes: Arc<RwLock<RateCounter>>) -> Result<(), Error> {
+	fn write(mut self, tracker: Arc<Tracker>) -> Result<(), Error> {
 		let mut msg = ser::ser_vec(&MsgHeader::new(self.resp_type, self.body.len() as u64))?;
 		msg.append(&mut self.body);
 		write_all(&mut self.stream, &msg[..], time::Duration::from_secs(10))?;
-		// Increase sent bytes counter
+		tracker.inc_sent(msg.len() as u64);
-		{
+
-			let mut sent_bytes = sent_bytes.write();
-			sent_bytes.inc(msg.len() as u64);
-		}
 		if let Some(mut file) = self.attachment {
 			let mut buf = [0u8; 8000];
 			loop {
@@ -148,8 +149,7 @@ impl<'a> Response<'a> {
 						write_all(&mut self.stream, &buf[..n], time::Duration::from_secs(10))?;
 						// Increase sent bytes "quietly" without incrementing the counter.
 						// (In a loop here for the single attachment).
-						let mut sent_bytes = sent_bytes.write();
+						tracker.inc_quiet_sent(n as u64);
-						sent_bytes.inc_quiet(n as u64);
 					}
 					Err(e) => return Err(From::from(e)),
 				}
@@ -165,72 +165,120 @@ impl<'a> Response<'a> {
 
 pub const SEND_CHANNEL_CAP: usize = 10;
 
-pub struct Tracker {
+pub struct StopHandle {
-	/// Bytes we've sent.
-	pub sent_bytes: Arc<RwLock<RateCounter>>,
-	/// Bytes we've received.
-	pub received_bytes: Arc<RwLock<RateCounter>>,
-	/// Channel to allow sending data through the connection
-	pub send_channel: mpsc::SyncSender<Vec<u8>>,
 	/// Channel to close the connection
 	pub close_channel: mpsc::Sender<()>,
+	// we need Option to take ownhership of the handle in stop()
+	peer_thread: Option<JoinHandle<()>>,
 }
 
-impl Tracker {
+impl StopHandle {
-	pub fn send<T>(&self, body: T, msg_type: Type) -> Result<(), Error>
+	/// Schedule this connection to safely close via the async close_channel.
+	pub fn stop(&self) {
+		if self.close_channel.send(()).is_err() {
+			debug!("peer's close_channel is disconnected, must be stopped already");
+			return;
+		}
+	}
+
+	pub fn stop_and_wait(&mut self) {
+		self.stop();
+		if let Some(peer_thread) = self.peer_thread.take() {
+			// wait only if other thread is calling us, eg shutdown
+			if thread::current().id() != peer_thread.thread().id() {
+				debug!("waiting for thread {:?} exit", peer_thread.thread().id());
+				if let Err(e) = peer_thread.join() {
+					error!("failed to stop peer thread: {:?}", e);
+				}
+			} else {
+				debug!(
+					"attempt to stop thread {:?} from itself",
+					peer_thread.thread().id()
+				);
+			}
+		}
+	}
+}
+
+pub struct ConnHandle {
+	/// Channel to allow sending data through the connection
+	pub send_channel: mpsc::SyncSender<Vec<u8>>,
+}
+
+impl ConnHandle {
+	pub fn send<T>(&self, body: T, msg_type: Type) -> Result<u64, Error>
 	where
 		T: ser::Writeable,
 	{
 		let buf = write_to_buf(body, msg_type)?;
 		let buf_len = buf.len();
 		self.send_channel.try_send(buf)?;
-
+		Ok(buf_len as u64)
-		// Increase sent bytes counter
+	}
-		let mut sent_bytes = self.sent_bytes.write();
-		sent_bytes.inc(buf_len as u64);
-
-		Ok(())
 }
 
-	/// Schedule this connection to safely close via the async close_channel.
+pub struct Tracker {
-	pub fn close(&self) {
+	/// Bytes we've sent.
-		let _ = self.close_channel.send(());
+	pub sent_bytes: Arc<RwLock<RateCounter>>,
+	/// Bytes we've received.
+	pub received_bytes: Arc<RwLock<RateCounter>>,
+}
+
+impl Tracker {
+	pub fn new() -> Tracker {
+		let received_bytes = Arc::new(RwLock::new(RateCounter::new()));
+		let sent_bytes = Arc::new(RwLock::new(RateCounter::new()));
+		Tracker {
+			received_bytes,
+			sent_bytes,
+		}
+	}
+
+	pub fn inc_received(&self, size: u64) {
+		self.received_bytes.write().inc(size);
+	}
+
+	pub fn inc_sent(&self, size: u64) {
+		self.sent_bytes.write().inc(size);
+	}
+
+	pub fn inc_quiet_received(&self, size: u64) {
+		self.received_bytes.write().inc_quiet(size);
+	}
+
+	pub fn inc_quiet_sent(&self, size: u64) {
+		self.sent_bytes.write().inc_quiet(size);
 	}
 }
 
 /// Start listening on the provided connection and wraps it. Does not hang
 /// the current thread, instead just returns a future and the Connection
 /// itself.
-pub fn listen<H>(stream: TcpStream, handler: H) -> Tracker
+pub fn listen<H>(
+	stream: TcpStream,
+	tracker: Arc<Tracker>,
+	handler: H,
+) -> io::Result<(ConnHandle, StopHandle)>
 where
 	H: MessageHandler,
 {
 	let (send_tx, send_rx) = mpsc::sync_channel(SEND_CHANNEL_CAP);
 	let (close_tx, close_rx) = mpsc::channel();
 
-	// Counter of number of bytes received
-	let received_bytes = Arc::new(RwLock::new(RateCounter::new()));
-	// Counter of number of bytes sent
-	let sent_bytes = Arc::new(RwLock::new(RateCounter::new()));
-
 	stream
 		.set_nonblocking(true)
 		.expect("Non-blocking IO not available.");
-	poll(
+	let peer_thread = poll(stream, handler, send_rx, close_rx, tracker)?;
-		stream,
-		handler,
-		send_rx,
-		close_rx,
-		received_bytes.clone(),
-		sent_bytes.clone(),
-	);
 
-	Tracker {
+	Ok((
-		sent_bytes: sent_bytes.clone(),
+		ConnHandle {
-		received_bytes: received_bytes.clone(),
 			send_channel: send_tx,
+		},
+		StopHandle {
 			close_channel: close_tx,
-	}
+			peer_thread: Some(peer_thread),
+		},
+	))
 }
 
 fn poll<H>(
@@ -238,16 +286,16 @@ fn poll<H>(
 	handler: H,
 	send_rx: mpsc::Receiver<Vec<u8>>,
 	close_rx: mpsc::Receiver<()>,
-	received_bytes: Arc<RwLock<RateCounter>>,
+	tracker: Arc<Tracker>,
-	sent_bytes: Arc<RwLock<RateCounter>>,
+) -> io::Result<JoinHandle<()>>
-) where
+where
 	H: MessageHandler,
 {
 	// Split out tcp stream out into separate reader/writer halves.
 	let mut reader = conn.try_clone().expect("clone conn for reader failed");
 	let mut writer = conn.try_clone().expect("clone conn for writer failed");
 
-	let _ = thread::Builder::new()
+	thread::Builder::new()
 		.name("peer".to_string())
 		.spawn(move || {
 			let sleep_time = time::Duration::from_millis(5);
@@ -265,19 +313,17 @@ fn poll<H>(
 						);
 
 						// Increase received bytes counter
-						received_bytes
+						tracker.inc_received(MsgHeader::LEN as u64 + msg.header.msg_len);
-							.write()
-							.inc(MsgHeader::LEN as u64 + msg.header.msg_len);
 
 						if let Some(Some(resp)) =
-							try_break!(handler.consume(msg, &mut writer, received_bytes.clone()))
+							try_break!(handler.consume(msg, &mut writer, tracker.clone()))
 						{
-							try_break!(resp.write(sent_bytes.clone()));
+							try_break!(resp.write(tracker.clone()));
 						}
 					}
 					Some(MsgHeaderWrapper::Unknown(msg_len)) => {
 						// Increase received bytes counter
-						received_bytes.write().inc(MsgHeader::LEN as u64 + msg_len);
+						tracker.inc_received(MsgHeader::LEN as u64 + msg_len);
 
 						try_break!(read_discard(msg_len, &mut reader));
 					}
@@ -288,7 +334,12 @@ fn poll<H>(
 				let maybe_data = retry_send.or_else(|_| send_rx.try_recv());
 				retry_send = Err(());
 				if let Ok(data) = maybe_data {
-					let written = try_break!(writer.write_all(&data[..]).map_err(&From::from));
+					let written = try_break!(write_all(
+						&mut writer,
+						&data[..],
+						std::time::Duration::from_secs(10)
+					)
+					.map_err(&From::from));
 					if written.is_none() {
 						retry_send = Ok(data);
 					}
@@ -309,5 +360,5 @@ fn poll<H>(
 					.unwrap_or("?".to_owned())
 			);
 			let _ = conn.shutdown(Shutdown::Both);
-		});
+		})
 }

p2p/src/peer.rs

@@ -54,7 +54,12 @@ pub struct Peer {
 	state: Arc<RwLock<State>>,
 	// set of all hashes known to this peer (so no need to send)
 	tracking_adapter: TrackingAdapter,
-	connection: Mutex<conn::Tracker>,
+	tracker: Arc<conn::Tracker>,
+	send_handle: Mutex<conn::ConnHandle>,
+	// we need a special lock for stop operation, can't reuse handle mutex for that
+	// because it may be locked by different reasons, so we should wait for that, close
+	// mutex can be taken only during shutdown, it happens once
+	stop_handle: Mutex<conn::StopHandle>,
 }
 
 impl fmt::Debug for Peer {
@@ -65,17 +70,22 @@ impl fmt::Debug for Peer {
 
 impl Peer {
 	// Only accept and connect can be externally used to build a peer
-	fn new(info: PeerInfo, conn: TcpStream, adapter: Arc<dyn NetAdapter>) -> Peer {
+	fn new(info: PeerInfo, conn: TcpStream, adapter: Arc<dyn NetAdapter>) -> std::io::Result<Peer> {
 		let state = Arc::new(RwLock::new(State::Connected));
 		let tracking_adapter = TrackingAdapter::new(adapter);
 		let handler = Protocol::new(Arc::new(tracking_adapter.clone()), info.clone());
-		let connection = Mutex::new(conn::listen(conn, handler));
+		let tracker = Arc::new(conn::Tracker::new());
-		Peer {
+		let (sendh, stoph) = conn::listen(conn, tracker.clone(), handler)?;
+		let send_handle = Mutex::new(sendh);
+		let stop_handle = Mutex::new(stoph);
+		Ok(Peer {
 			info,
 			state,
 			tracking_adapter,
-			connection,
+			tracker,
-		}
+			send_handle,
+			stop_handle,
+		})
 	}
 
 	pub fn accept(
@@ -88,7 +98,7 @@ impl Peer {
 		debug!("accept: handshaking from {:?}", conn.peer_addr());
 		let info = hs.accept(capab, total_difficulty, &mut conn);
 		match info {
-			Ok(info) => Ok(Peer::new(info, conn, adapter)),
+			Ok(info) => Ok(Peer::new(info, conn, adapter)?),
 			Err(e) => {
 				debug!(
 					"accept: handshaking from {:?} failed with error: {:?}",
@@ -114,7 +124,7 @@ impl Peer {
 		debug!("connect: handshaking with {:?}", conn.peer_addr());
 		let info = hs.initiate(capab, total_difficulty, self_addr, &mut conn);
 		match info {
-			Ok(info) => Ok(Peer::new(info, conn, adapter)),
+			Ok(info) => Ok(Peer::new(info, conn, adapter)?),
 			Err(e) => {
 				debug!(
 					"connect: handshaking with {:?} failed with error: {:?}",
@@ -184,30 +194,26 @@ impl Peer {
 
 	/// Whether the peer is considered abusive, mostly for spammy nodes
 	pub fn is_abusive(&self) -> bool {
-		let conn = self.connection.lock();
+		let rec = self.tracker.received_bytes.read();
-		let rec = conn.received_bytes.read();
+		let sent = self.tracker.sent_bytes.read();
-		let sent = conn.sent_bytes.read();
 		rec.count_per_min() > MAX_PEER_MSG_PER_MIN || sent.count_per_min() > MAX_PEER_MSG_PER_MIN
 	}
 
 	/// Number of bytes sent to the peer
 	pub fn last_min_sent_bytes(&self) -> Option<u64> {
-		let conn = self.connection.lock();
+		let sent_bytes = self.tracker.sent_bytes.read();
-		let sent_bytes = conn.sent_bytes.read();
 		Some(sent_bytes.bytes_per_min())
 	}
 
 	/// Number of bytes received from the peer
 	pub fn last_min_received_bytes(&self) -> Option<u64> {
-		let conn = self.connection.lock();
+		let received_bytes = self.tracker.received_bytes.read();
-		let received_bytes = conn.received_bytes.read();
 		Some(received_bytes.bytes_per_min())
 	}
 
 	pub fn last_min_message_counts(&self) -> Option<(u64, u64)> {
-		let conn = self.connection.lock();
+		let received_bytes = self.tracker.received_bytes.read();
-		let received_bytes = conn.received_bytes.read();
+		let sent_bytes = self.tracker.sent_bytes.read();
-		let sent_bytes = conn.sent_bytes.read();
 		Some((sent_bytes.count_per_min(), received_bytes.count_per_min()))
 	}
 
@@ -218,7 +224,9 @@ impl Peer {
 
 	/// Send a msg with given msg_type to our peer via the connection.
 	fn send<T: Writeable>(&self, msg: T, msg_type: Type) -> Result<(), Error> {
-		self.connection.lock().send(msg, msg_type)
+		let bytes = self.send_handle.lock().send(msg, msg_type)?;
+		self.tracker.inc_sent(bytes);
+		Ok(())
 	}
 
 	/// Send a ping to the remote peer, providing our local difficulty and
@@ -384,14 +392,25 @@ impl Peer {
 
 	pub fn send_kernel_data_request(&self) -> Result<(), Error> {
 		debug!("Asking {} for kernel data.", self.info.addr);
-		self.connection
+		self.send(&KernelDataRequest {}, msg::Type::KernelDataRequest)
-			.lock()
-			.send(&KernelDataRequest {}, msg::Type::KernelDataRequest)
 	}
 
-	/// Stops the peer, closing its connection
+	/// Stops the peer
 	pub fn stop(&self) {
-		self.connection.lock().close();
+		debug!("Stopping peer without waiting {:?}", self.info.addr);
+		match self.stop_handle.try_lock() {
+			Some(handle) => handle.stop(),
+			None => error!("can't get stop lock for peer"),
+		}
+	}
+
+	/// Stops the peer and wait until peer's thread exit
+	pub fn stop_and_wait(&self) {
+		debug!("Stopping peer {:?}", self.info.addr);
+		match self.stop_handle.try_lock() {
+			Some(mut handle) => handle.stop_and_wait(),
+			None => error!("can't get stop lock for peer"),
+		}
 	}
 }
 

p2p/src/peers.rs

@@ -27,15 +27,16 @@ use crate::core::core;
 use crate::core::core::hash::{Hash, Hashed};
 use crate::core::global;
 use crate::core::pow::Difficulty;
-use chrono::prelude::*;
-use chrono::Duration;
-
 use crate::peer::Peer;
 use crate::store::{PeerData, PeerStore, State};
 use crate::types::{
 	Capabilities, ChainAdapter, Error, NetAdapter, P2PConfig, PeerAddr, PeerInfo, ReasonForBan,
 	TxHashSetRead, MAX_PEER_ADDRS,
 };
+use chrono::prelude::*;
+use chrono::Duration;
+
+const LOCK_TIMEOUT: std::time::Duration = std::time::Duration::from_secs(2);
 
 pub struct Peers {
 	pub adapter: Arc<dyn ChainAdapter>,
@@ -57,6 +58,13 @@ impl Peers {
 	/// Adds the peer to our internal peer mapping. Note that the peer is still
 	/// returned so the server can run it.
 	pub fn add_connected(&self, peer: Arc<Peer>) -> Result<(), Error> {
+		let mut peers = match self.peers.try_write_for(LOCK_TIMEOUT) {
+			Some(peers) => peers,
+			None => {
+				error!("add_connected: failed to get peers lock");
+				return Err(Error::Timeout);
+			}
+		};
 		let peer_data = PeerData {
 			addr: peer.info.addr,
 			capabilities: peer.info.capabilities,
@@ -68,7 +76,7 @@ impl Peers {
 		};
 		debug!("Saving newly connected peer {}.", peer_data.addr);
 		self.save_peer(&peer_data)?;
-		self.peers.write().insert(peer_data.addr, peer.clone());
+		peers.insert(peer_data.addr, peer.clone());
 
 		Ok(())
 	}
@@ -90,14 +98,26 @@ impl Peers {
 	}
 
 	pub fn is_known(&self, addr: PeerAddr) -> bool {
-		self.peers.read().contains_key(&addr)
+		let peers = match self.peers.try_read_for(LOCK_TIMEOUT) {
+			Some(peers) => peers,
+			None => {
+				error!("is_known: failed to get peers lock");
+				return false;
+			}
+		};
+		peers.contains_key(&addr)
 	}
 
 	/// Get vec of peers we are currently connected to.
 	pub fn connected_peers(&self) -> Vec<Arc<Peer>> {
-		let mut res = self
+		let peers = match self.peers.try_read_for(LOCK_TIMEOUT) {
-			.peers
+			Some(peers) => peers,
-			.read()
+			None => {
+				error!("connected_peers: failed to get peers lock");
+				return vec![];
+			}
+		};
+		let mut res = peers
 			.values()
 			.filter(|p| p.is_connected())
 			.cloned()
@@ -115,7 +135,14 @@ impl Peers {
 
 	/// Get a peer we're connected to by address.
 	pub fn get_connected_peer(&self, addr: PeerAddr) -> Option<Arc<Peer>> {
-		self.peers.read().get(&addr).map(|p| p.clone())
+		let peers = match self.peers.try_read_for(LOCK_TIMEOUT) {
+			Some(peers) => peers,
+			None => {
+				error!("get_connected_peer: failed to get peers lock");
+				return None;
+			}
+		};
+		peers.get(&addr).map(|p| p.clone())
 	}
 
 	/// Number of peers currently connected to.
@@ -204,9 +231,7 @@ impl Peers {
 
 	pub fn is_banned(&self, peer_addr: PeerAddr) -> bool {
 		if let Ok(peer) = self.store.get_peer(peer_addr) {
-			if peer.flags == State::Banned {
+			return peer.flags == State::Banned;
-				return true;
-			}
 		}
 		false
 	}
@@ -215,6 +240,7 @@ impl Peers {
 	pub fn ban_peer(&self, peer_addr: PeerAddr, ban_reason: ReasonForBan) {
 		if let Err(e) = self.update_state(peer_addr, State::Banned) {
 			error!("Couldn't ban {}: {:?}", peer_addr, e);
+			return;
 		}
 
 		if let Some(peer) = self.get_connected_peer(peer_addr) {
@@ -226,7 +252,15 @@ impl Peers {
 			};
 			peer.set_banned();
 			peer.stop();
-			self.peers.write().remove(&peer.info.addr);
+
+			let mut peers = match self.peers.try_write_for(LOCK_TIMEOUT) {
+				Some(peers) => peers,
+				None => {
+					error!("ban_peer: failed to get peers lock");
+					return;
+				}
+			};
+			peers.remove(&peer.info.addr);
 		}
 	}
 
@@ -264,8 +298,16 @@ impl Peers {
 						"Error sending {:?} to peer {:?}: {:?}",
 						obj_name, &p.info.addr, e
 					);
+
+					let mut peers = match self.peers.try_write_for(LOCK_TIMEOUT) {
+						Some(peers) => peers,
+						None => {
+							error!("broadcast: failed to get peers lock");
+							break;
+						}
+					};
 					p.stop();
-					self.peers.write().remove(&p.info.addr);
+					peers.remove(&p.info.addr);
 				}
 			}
 
@@ -331,8 +373,15 @@ impl Peers {
 		for p in self.connected_peers().iter() {
 			if let Err(e) = p.send_ping(total_difficulty, height) {
 				debug!("Error pinging peer {:?}: {:?}", &p.info.addr, e);
+				let mut peers = match self.peers.try_write_for(LOCK_TIMEOUT) {
+					Some(peers) => peers,
+					None => {
+						error!("check_all: failed to get peers lock");
+						break;
+					}
+				};
 				p.stop();
-				self.peers.write().remove(&p.info.addr);
+				peers.remove(&p.info.addr);
 			}
 		}
 	}
@@ -388,7 +437,15 @@ impl Peers {
 		let mut rm = vec![];
 
 		// build a list of peers to be cleaned up
-		for peer in self.peers.read().values() {
+		{
+			let peers = match self.peers.try_read_for(LOCK_TIMEOUT) {
+				Some(peers) => peers,
+				None => {
+					error!("clean_peers: can't get peers lock");
+					return;
+				}
+			};
+			for peer in peers.values() {
 				if peer.is_banned() {
 					debug!("clean_peers {:?}, peer banned", peer.info.addr);
 					rm.push(peer.info.addr.clone());
@@ -418,6 +475,7 @@ impl Peers {
 					}
 				}
 			}
+		}
 
 		// ensure we do not still have too many connected peers
 		let excess_count = (self.peer_count() as usize)
@@ -436,7 +494,13 @@ impl Peers {
 
 		// now clean up peer map based on the list to remove
 		{
-			let mut peers = self.peers.write();
+			let mut peers = match self.peers.try_write_for(LOCK_TIMEOUT) {
+				Some(peers) => peers,
+				None => {
+					error!("clean_peers: failed to get peers lock");
+					return;
+				}
+			};
 			for addr in rm {
 				let _ = peers.get(&addr).map(|peer| peer.stop());
 				peers.remove(&addr);
@@ -447,7 +511,7 @@ impl Peers {
 	pub fn stop(&self) {
 		let mut peers = self.peers.write();
 		for (_, peer) in peers.drain() {
-			peer.stop();
+			peer.stop_and_wait();
 		}
 	}
 

p2p/src/protocol.rs

@@ -12,24 +12,20 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use rand::{thread_rng, Rng};
+use crate::conn::{Message, MessageHandler, Response, Tracker};
-use std::cmp;
-use std::fs::{self, File, OpenOptions};
-use std::io::{BufWriter, Seek, SeekFrom, Write};
-use std::sync::Arc;
-
-use chrono::prelude::Utc;
-use tempfile::tempfile;
-
-use crate::conn::{Message, MessageHandler, Response};
 use crate::core::core::{self, hash::Hash, CompactBlock};
-use crate::util::{RateCounter, RwLock};
-
 use crate::msg::{
 	BanReason, GetPeerAddrs, Headers, KernelDataResponse, Locator, PeerAddrs, Ping, Pong,
 	TxHashSetArchive, TxHashSetRequest, Type,
 };
 use crate::types::{Error, NetAdapter, PeerInfo};
+use chrono::prelude::Utc;
+use rand::{thread_rng, Rng};
+use std::cmp;
+use std::fs::{self, File, OpenOptions};
+use std::io::{BufWriter, Seek, SeekFrom, Write};
+use std::sync::Arc;
+use tempfile::tempfile;
 
 pub struct Protocol {
 	adapter: Arc<dyn NetAdapter>,
@@ -47,7 +43,7 @@ impl MessageHandler for Protocol {
 		&self,
 		mut msg: Message<'a>,
 		writer: &'a mut dyn Write,
-		received_bytes: Arc<RwLock<RateCounter>>,
+		tracker: Arc<Tracker>,
 	) -> Result<Option<Response<'a>>, Error> {
 		let adapter = &self.adapter;
 
@@ -275,7 +271,7 @@ impl MessageHandler for Protocol {
 
 					// Increase received bytes quietly (without affecting the counters).
 					// Otherwise we risk banning a peer as "abusive".
-					received_bytes.write().inc_quiet(size as u64);
+					tracker.inc_quiet_received(size as u64);
 				}
 
 				// Remember to seek back to start of the file as the caller is likely
@@ -362,10 +358,7 @@ impl MessageHandler for Protocol {
 
 						// Increase received bytes quietly (without affecting the counters).
 						// Otherwise we risk banning a peer as "abusive".
-						{
+						tracker.inc_quiet_received(size as u64)
-							let mut received_bytes = received_bytes.write();
-							received_bytes.inc_quiet(size as u64);
-						}
 					}
 					tmp_zip
 						.into_inner()

p2p/src/serv.rs

@@ -33,7 +33,7 @@ use crate::types::{
 	Capabilities, ChainAdapter, Error, NetAdapter, P2PConfig, PeerAddr, PeerInfo, ReasonForBan,
 	TxHashSetRead,
 };
-use crate::util::{Mutex, StopState};
+use crate::util::StopState;
 use chrono::prelude::{DateTime, Utc};
 
 /// P2P server implementation, handling bootstrapping to find and connect to
@@ -43,7 +43,7 @@ pub struct Server {
 	capabilities: Capabilities,
 	handshake: Arc<Handshake>,
 	pub peers: Arc<Peers>,
-	stop_state: Arc<Mutex<StopState>>,
+	stop_state: Arc<StopState>,
 }
 
 // TODO TLS
@@ -55,7 +55,7 @@ impl Server {
 		config: P2PConfig,
 		adapter: Arc<dyn ChainAdapter>,
 		genesis: Hash,
-		stop_state: Arc<Mutex<StopState>>,
+		stop_state: Arc<StopState>,
 	) -> Result<Server, Error> {
 		Ok(Server {
 			config: config.clone(),
@@ -77,7 +77,7 @@ impl Server {
 		let sleep_time = Duration::from_millis(5);
 		loop {
 			// Pause peer ingress connection request. Only for tests.
-			if self.stop_state.lock().is_paused() {
+			if self.stop_state.is_paused() {
 				thread::sleep(Duration::from_secs(1));
 				continue;
 			}
@@ -89,10 +89,14 @@ impl Server {
 					if self.check_undesirable(&stream) {
 						continue;
 					}
-					if let Err(e) = self.handle_new_peer(stream) {
+					match self.handle_new_peer(stream) {
+						Err(Error::ConnectionClose) => debug!("shutting down, ignoring a new peer"),
+						Err(e) => {
 							debug!("Error accepting peer {}: {:?}", peer_addr.to_string(), e);
 							let _ = self.peers.add_banned(peer_addr, ReasonForBan::BadHandshake);
 						}
+						Ok(_) => {}
+					}
 				}
 				Err(ref e) if e.kind() == io::ErrorKind::WouldBlock => {
 					// nothing to do, will retry in next iteration
@@ -101,7 +105,7 @@ impl Server {
 					debug!("Couldn't establish new client connection: {:?}", e);
 				}
 			}
-			if self.stop_state.lock().is_stopped() {
+			if self.stop_state.is_stopped() {
 				break;
 			}
 			thread::sleep(sleep_time);
@@ -112,6 +116,10 @@ impl Server {
 	/// Asks the server to connect to a new peer. Directly returns the peer if
 	/// we're already connected to the provided address.
 	pub fn connect(&self, addr: PeerAddr) -> Result<Arc<Peer>, Error> {
+		if self.stop_state.is_stopped() {
+			return Err(Error::ConnectionClose);
+		}
+
 		if Peer::is_denied(&self.config, addr) {
 			debug!("connect_peer: peer {} denied, not connecting.", addr);
 			return Err(Error::ConnectionClose);
@@ -169,6 +177,9 @@ impl Server {
 	}
 
 	fn handle_new_peer(&self, stream: TcpStream) -> Result<(), Error> {
+		if self.stop_state.is_stopped() {
+			return Err(Error::ConnectionClose);
+		}
 		let total_diff = self.peers.total_difficulty()?;
 
 		// accept the peer and add it to the server map
@@ -214,7 +225,7 @@ impl Server {
 	}
 
 	pub fn stop(&self) {
-		self.stop_state.lock().stop();
+		self.stop_state.stop();
 		self.peers.stop();
 	}
 

p2p/tests/peer_handshake.rs

@@ -16,7 +16,7 @@ use grin_core as core;
 use grin_p2p as p2p;
 
 use grin_util as util;
-use grin_util::{Mutex, StopState};
+use grin_util::StopState;
 
 use std::net::{SocketAddr, TcpListener, TcpStream};
 use std::sync::Arc;
@@ -56,7 +56,7 @@ fn peer_handshake() {
 			p2p_config.clone(),
 			net_adapter.clone(),
 			Hash::from_vec(&vec![]),
-			Arc::new(Mutex::new(StopState::new())),
+			Arc::new(StopState::new()),
 		)
 		.unwrap(),
 	);

servers/src/common/types.rs

@@ -344,6 +344,7 @@ pub enum SyncStatus {
 		current_height: u64,
 		highest_height: u64,
 	},
+	Shutdown,
 }
 
 /// Current sync state. Encapsulates the current SyncStatus.

servers/src/grin/dandelion_monitor.rs

@@ -16,14 +16,14 @@ use chrono::prelude::Utc;
 use rand::{thread_rng, Rng};
 use std::sync::Arc;
 use std::thread;
-use std::time::Duration;
+use std::time::{Duration, Instant};
 
 use crate::common::adapters::DandelionAdapter;
 use crate::core::core::hash::Hashed;
 use crate::core::core::transaction;
 use crate::core::core::verifier_cache::VerifierCache;
 use crate::pool::{DandelionConfig, Pool, PoolEntry, PoolError, TransactionPool, TxSource};
-use crate::util::{Mutex, RwLock, StopState};
+use crate::util::{RwLock, StopState};
 
 /// A process to monitor transactions in the stempool.
 /// With Dandelion, transaction can be broadcasted in stem or fluff phase.
@@ -38,22 +38,31 @@ pub fn monitor_transactions(
 	tx_pool: Arc<RwLock<TransactionPool>>,
 	adapter: Arc<DandelionAdapter>,
 	verifier_cache: Arc<RwLock<dyn VerifierCache>>,
-	stop_state: Arc<Mutex<StopState>>,
+	stop_state: Arc<StopState>,
-) {
+) -> std::io::Result<thread::JoinHandle<()>> {
 	debug!("Started Dandelion transaction monitor.");
 
-	let _ = thread::Builder::new()
+	thread::Builder::new()
 		.name("dandelion".to_string())
 		.spawn(move || {
+			let run_interval = Duration::from_secs(10);
+			let mut last_run = Instant::now()
+				.checked_sub(Duration::from_secs(20))
+				.unwrap_or_else(|| Instant::now());
 			loop {
 				// Halt Dandelion monitor if we have been notified that we are stopping.
-				if stop_state.lock().is_stopped() {
+				if stop_state.is_stopped() {
 					break;
 				}
 
+				if last_run.elapsed() > run_interval {
 					if !adapter.is_stem() {
-					let _ =
+						let _ = process_fluff_phase(
-						process_fluff_phase(&dandelion_config, &tx_pool, &adapter, &verifier_cache)
+							&dandelion_config,
+							&tx_pool,
+							&adapter,
+							&verifier_cache,
+						)
 						.map_err(|e| {
 							error!("dand_mon: Problem processing fluff phase. {:?}", e);
 						});
@@ -70,11 +79,13 @@ pub fn monitor_transactions(
 					if adapter.is_expired() {
 						adapter.next_epoch();
 					}
-
+					last_run = Instant::now();
-				// Monitor loops every 10s.
-				thread::sleep(Duration::from_secs(10));
 				}
-		});
+
+				// Monitor loops every 10s, but check stop flag every second.
+				thread::sleep(Duration::from_secs(1));
+			}
+		})
 }
 
 // Query the pool for transactions older than the cutoff.

servers/src/grin/seed.rs

@@ -30,7 +30,7 @@ use crate::core::global;
 use crate::p2p;
 use crate::p2p::types::PeerAddr;
 use crate::p2p::ChainAdapter;
-use crate::util::{Mutex, StopState};
+use crate::util::StopState;
 
 // DNS Seeds with contact email associated
 const MAINNET_DNS_SEEDS: &'static [&'static str] = &[
@@ -54,9 +54,9 @@ pub fn connect_and_monitor(
 	capabilities: p2p::Capabilities,
 	seed_list: Box<dyn Fn() -> Vec<PeerAddr> + Send>,
 	preferred_peers: Option<Vec<PeerAddr>>,
-	stop_state: Arc<Mutex<StopState>>,
+	stop_state: Arc<StopState>,
-) {
+) -> std::io::Result<thread::JoinHandle<()>> {
-	let _ = thread::Builder::new()
+	thread::Builder::new()
 		.name("seed".to_string())
 		.spawn(move || {
 			let peers = p2p_server.peers.clone();
@@ -77,16 +77,15 @@ pub fn connect_and_monitor(
 			let mut prev_expire_check = MIN_DATE.and_hms(0, 0, 0);
 			let mut prev_ping = Utc::now();
 			let mut start_attempt = 0;
-
 			let mut connecting_history: HashMap<PeerAddr, DateTime<Utc>> = HashMap::new();
 
 			loop {
-				if stop_state.lock().is_stopped() {
+				if stop_state.is_stopped() {
 					break;
 				}
 
 				// Pause egress peer connection request. Only for tests.
-				if stop_state.lock().is_paused() {
+				if stop_state.is_paused() {
 					thread::sleep(time::Duration::from_secs(1));
 					continue;
 				}
@@ -136,7 +135,7 @@ pub fn connect_and_monitor(
 
 				thread::sleep(time::Duration::from_secs(1));
 			}
-		});
+		})
 }
 
 fn monitor_peers(
@@ -326,17 +325,19 @@ fn listen_for_addrs(
 
 		let peers_c = peers.clone();
 		let p2p_c = p2p.clone();
-		let _ = thread::Builder::new()
+		thread::Builder::new()
 			.name("peer_connect".to_string())
 			.spawn(move || match p2p_c.connect(addr) {
 				Ok(p) => {
-					let _ = p.send_peer_request(capab);
+					if p.send_peer_request(capab).is_ok() {
 						let _ = peers_c.update_state(addr, p2p::State::Healthy);
 					}
+				}
 				Err(_) => {
 					let _ = peers_c.update_state(addr, p2p::State::Defunct);
 				}
-			});
+			})
+			.expect("failed to launch peer_connect thread");
 	}
 
 	// shrink the connecting history.

servers/src/grin/server.rs

@@ -21,7 +21,10 @@ use std::fs::File;
 use std::io::prelude::*;
 use std::path::Path;
 use std::sync::Arc;
-use std::{thread, time};
+use std::{
+	thread::{self, JoinHandle},
+	time,
+};
 
 use fs2::FileExt;
 
@@ -44,7 +47,7 @@ use crate::p2p;
 use crate::p2p::types::PeerAddr;
 use crate::pool;
 use crate::util::file::get_first_line;
-use crate::util::{Mutex, RwLock, StopState};
+use crate::util::{RwLock, StopState};
 
 /// Grin server holding internal structures.
 pub struct Server {
@@ -64,9 +67,13 @@ pub struct Server {
 	/// To be passed around to collect stats and info
 	state_info: ServerStateInfo,
 	/// Stop flag
-	pub stop_state: Arc<Mutex<StopState>>,
+	pub stop_state: Arc<StopState>,
 	/// Maintain a lock_file so we do not run multiple Grin nodes from same dir.
 	lock_file: Arc<File>,
+	connect_thread: Option<JoinHandle<()>>,
+	sync_thread: JoinHandle<()>,
+	dandelion_thread: JoinHandle<()>,
+	p2p_thread: JoinHandle<()>,
 }
 
 impl Server {
@@ -142,7 +149,7 @@ impl Server {
 			Some(b) => b,
 		};
 
-		let stop_state = Arc::new(Mutex::new(StopState::new()));
+		let stop_state = Arc::new(StopState::new());
 
 		// Shared cache for verification results.
 		// We cache rangeproof verification and kernel signature verification.
@@ -180,7 +187,6 @@ impl Server {
 			pow::verify_size,
 			verifier_cache.clone(),
 			archive_mode,
-			stop_state.clone(),
 		)?);
 
 		pool_adapter.set_chain(shared_chain.clone());
@@ -208,6 +214,8 @@ impl Server {
 		pool_net_adapter.init(p2p_server.peers.clone());
 		net_adapter.init(p2p_server.peers.clone());
 
+		let mut connect_thread = None;
+
 		if config.p2p_config.seeding_type != p2p::Seeding::Programmatic {
 			let seeder = match config.p2p_config.seeding_type {
 				p2p::Seeding::None => {
@@ -226,13 +234,13 @@ impl Server {
 				_ => unreachable!(),
 			};
 
-			seed::connect_and_monitor(
+			connect_thread = Some(seed::connect_and_monitor(
 				p2p_server.clone(),
 				config.p2p_config.capabilities,
 				seeder,
 				config.p2p_config.peers_preferred.clone(),
 				stop_state.clone(),
-			);
+			)?);
 		}
 
 		// Defaults to None (optional) in config file.
@@ -240,17 +248,21 @@ impl Server {
 		let skip_sync_wait = config.skip_sync_wait.unwrap_or(false);
 		sync_state.update(SyncStatus::AwaitingPeers(!skip_sync_wait));
 
-		sync::run_sync(
+		let sync_thread = sync::run_sync(
 			sync_state.clone(),
 			p2p_server.peers.clone(),
 			shared_chain.clone(),
 			stop_state.clone(),
-		);
+		)?;
 
 		let p2p_inner = p2p_server.clone();
-		let _ = thread::Builder::new()
+		let p2p_thread = thread::Builder::new()
 			.name("p2p-server".to_string())
-			.spawn(move || p2p_inner.listen());
+			.spawn(move || {
+				if let Err(e) = p2p_inner.listen() {
+					error!("P2P server failed with erorr: {:?}", e);
+				}
+			})?;
 
 		info!("Starting rest apis at: {}", &config.api_http_addr);
 		let api_secret = get_first_line(config.api_secret_path.clone());
@@ -269,6 +281,7 @@ impl Server {
 			}
 		};
 
+		// TODO fix API shutdown and join this thread
 		api::start_rest_apis(
 			config.api_http_addr.clone(),
 			shared_chain.clone(),
@@ -279,13 +292,13 @@ impl Server {
 		);
 
 		info!("Starting dandelion monitor: {}", &config.api_http_addr);
-		dandelion_monitor::monitor_transactions(
+		let dandelion_thread = dandelion_monitor::monitor_transactions(
 			config.dandelion_config.clone(),
 			tx_pool.clone(),
 			pool_net_adapter.clone(),
 			verifier_cache.clone(),
 			stop_state.clone(),
-		);
+		)?;
 
 		warn!("Grin server started.");
 		Ok(Server {
@@ -300,6 +313,10 @@ impl Server {
 			},
 			stop_state,
 			lock_file,
+			connect_thread,
+			sync_thread,
+			p2p_thread,
+			dandelion_thread,
 		})
 	}
 
@@ -348,7 +365,7 @@ impl Server {
 	pub fn start_test_miner(
 		&self,
 		wallet_listener_url: Option<String>,
-		stop_state: Arc<Mutex<StopState>>,
+		stop_state: Arc<StopState>,
 	) {
 		info!("start_test_miner - start",);
 		let sync_state = self.sync_state.clone();
@@ -489,15 +506,41 @@ impl Server {
 	}
 
 	/// Stop the server.
-	pub fn stop(&self) {
+	pub fn stop(self) {
+		{
+			self.sync_state.update(SyncStatus::Shutdown);
+			self.stop_state.stop();
+
+			if let Some(connect_thread) = self.connect_thread {
+				match connect_thread.join() {
+					Err(e) => error!("failed to join to connect_and_monitor thread: {:?}", e),
+					Ok(_) => info!("connect_and_monitor thread stopped"),
+				}
+			} else {
+				info!("No active connect_and_monitor thread")
+			}
+
+			match self.sync_thread.join() {
+				Err(e) => error!("failed to join to sync thread: {:?}", e),
+				Ok(_) => info!("sync thread stopped"),
+			}
+
+			match self.dandelion_thread.join() {
+				Err(e) => error!("failed to join to dandelion_monitor thread: {:?}", e),
+				Ok(_) => info!("dandelion_monitor thread stopped"),
+			}
+		}
 		self.p2p.stop();
-		self.stop_state.lock().stop();
+		match self.p2p_thread.join() {
+			Err(e) => error!("failed to join to p2p thread: {:?}", e),
+			Ok(_) => info!("p2p thread stopped"),
+		}
 		let _ = self.lock_file.unlock();
 	}
 
 	/// Pause the p2p server.
 	pub fn pause(&self) {
-		self.stop_state.lock().pause();
+		self.stop_state.pause();
 		thread::sleep(time::Duration::from_secs(1));
 		self.p2p.pause();
 	}
@@ -505,12 +548,12 @@ impl Server {
 	/// Resume p2p server.
 	/// TODO - We appear not to resume the p2p server (peer connections) here?
 	pub fn resume(&self) {
-		self.stop_state.lock().resume();
+		self.stop_state.resume();
 	}
 
 	/// Stops the test miner without stopping the p2p layer
-	pub fn stop_test_miner(&self, stop: Arc<Mutex<StopState>>) {
+	pub fn stop_test_miner(&self, stop: Arc<StopState>) {
-		stop.lock().stop();
+		stop.stop();
 		info!("stop_test_miner - stop",);
 	}
 }

servers/src/grin/sync/body_sync.rs

@@ -141,6 +141,7 @@ impl BodySync {
 				if let Some(peer) = peers_iter.next() {
 					if let Err(e) = peer.send_block_request(*hash) {
 						debug!("Skipped request to {}: {:?}", peer.info.addr, e);
+						peer.stop();
 					} else {
 						self.blocks_requested += 1;
 					}

servers/src/grin/sync/syncer.rs

@@ -24,27 +24,27 @@ use crate::grin::sync::body_sync::BodySync;
 use crate::grin::sync::header_sync::HeaderSync;
 use crate::grin::sync::state_sync::StateSync;
 use crate::p2p;
-use crate::util::{Mutex, StopState};
+use crate::util::StopState;
 
 pub fn run_sync(
 	sync_state: Arc<SyncState>,
 	peers: Arc<p2p::Peers>,
 	chain: Arc<chain::Chain>,
-	stop_state: Arc<Mutex<StopState>>,
+	stop_state: Arc<StopState>,
-) {
+) -> std::io::Result<std::thread::JoinHandle<()>> {
-	let _ = thread::Builder::new()
+	thread::Builder::new()
 		.name("sync".to_string())
 		.spawn(move || {
 			let runner = SyncRunner::new(sync_state, peers, chain, stop_state);
 			runner.sync_loop();
-		});
+		})
 }
 
 pub struct SyncRunner {
 	sync_state: Arc<SyncState>,
 	peers: Arc<p2p::Peers>,
 	chain: Arc<chain::Chain>,
-	stop_state: Arc<Mutex<StopState>>,
+	stop_state: Arc<StopState>,
 }
 
 impl SyncRunner {
@@ -52,7 +52,7 @@ impl SyncRunner {
 		sync_state: Arc<SyncState>,
 		peers: Arc<p2p::Peers>,
 		chain: Arc<chain::Chain>,
-		stop_state: Arc<Mutex<StopState>>,
+		stop_state: Arc<StopState>,
 	) -> SyncRunner {
 		SyncRunner {
 			sync_state,
@@ -77,6 +77,9 @@ impl SyncRunner {
 		let mut n = 0;
 		const MIN_PEERS: usize = 3;
 		loop {
+			if self.stop_state.is_stopped() {
+				break;
+			}
 			let wp = self.peers.more_or_same_work_peers()?;
 			// exit loop when:
 			// * we have more than MIN_PEERS more_or_same_work peers
@@ -140,7 +143,7 @@ impl SyncRunner {
 
 		// Main syncing loop
 		loop {
-			if self.stop_state.lock().is_stopped() {
+			if self.stop_state.is_stopped() {
 				break;
 			}
 
@@ -167,7 +170,13 @@ impl SyncRunner {
 					unwrap_or_restart_loop!(self.chain.compact());
 				}
 
-				thread::sleep(time::Duration::from_secs(10));
+				// sleep for 10 secs but check stop signal every second
+				for _ in 1..10 {
+					thread::sleep(time::Duration::from_secs(1));
+					if self.stop_state.is_stopped() {
+						break;
+					}
+				}
 				continue;
 			}
 

servers/src/mining/test_miner.rs

@@ -29,14 +29,14 @@ use crate::core::core::{Block, BlockHeader};
 use crate::core::global;
 use crate::mining::mine_block;
 use crate::pool;
-use crate::util::{Mutex, StopState};
+use crate::util::StopState;
 
 pub struct Miner {
 	config: StratumServerConfig,
 	chain: Arc<chain::Chain>,
 	tx_pool: Arc<RwLock<pool::TransactionPool>>,
 	verifier_cache: Arc<RwLock<dyn VerifierCache>>,
-	stop_state: Arc<Mutex<StopState>>,
+	stop_state: Arc<StopState>,
 
 	// Just to hold the port we're on, so this miner can be identified
 	// while watching debug output
@@ -51,7 +51,7 @@ impl Miner {
 		chain: Arc<chain::Chain>,
 		tx_pool: Arc<RwLock<pool::TransactionPool>>,
 		verifier_cache: Arc<RwLock<dyn VerifierCache>>,
-		stop_state: Arc<Mutex<StopState>>,
+		stop_state: Arc<StopState>,
 	) -> Miner {
 		Miner {
 			config,
@@ -136,7 +136,7 @@ impl Miner {
 		let mut key_id = None;
 
 		loop {
-			if self.stop_state.lock().is_stopped() {
+			if self.stop_state.is_stopped() {
 				break;
 			}
 

src/bin/tui/status.rs

@@ -199,6 +199,7 @@ impl TUIStatusListener for TUIStatusView {
 					};
 					format!("Downloading blocks: {}%, step 4/4", percent)
 				}
+				SyncStatus::Shutdown => "Shutting down, closing connections".to_string(),
 			}
 		};
 		/*let basic_mining_config_status = {

src/bin/tui/ui.rs

@@ -173,8 +173,10 @@ impl Controller {
 			while let Some(message) = self.rx.try_iter().next() {
 				match message {
 					ControllerMessage::Shutdown => {
-						server.stop();
 						self.ui.stop();
+						println!("Shutdown in progress, please wait");
+						server.stop();
+						return;
 					}
 				}
 			}

util/src/lib.rs

@@ -53,8 +53,8 @@ pub mod read_write;
 // other utils
 #[allow(unused_imports)]
 use std::ops::Deref;
+use std::sync::atomic::{AtomicBool, Ordering};
 use std::sync::Arc;
-
 mod hex;
 pub use crate::hex::*;
 
@@ -112,49 +112,45 @@ pub fn to_base64(s: &str) -> String {
 
 /// Global stopped/paused state shared across various subcomponents of Grin.
 ///
-/// Arc<Mutex<StopState>> allows the chain to lock the stop_state during critical processing.
-/// Other subcomponents cannot abruptly shutdown the server during block/header processing.
-/// This should prevent the chain ever ending up in an inconsistent state on restart.
-///
 /// "Stopped" allows a clean shutdown of the Grin server.
 /// "Paused" is used in some tests to allow nodes to reach steady state etc.
 ///
 pub struct StopState {
-	stopped: bool,
+	stopped: AtomicBool,
-	paused: bool,
+	paused: AtomicBool,
 }
 
 impl StopState {
 	/// Create a new stop_state in default "running" state.
 	pub fn new() -> StopState {
 		StopState {
-			stopped: false,
+			stopped: AtomicBool::new(false),
-			paused: false,
+			paused: AtomicBool::new(false),
 		}
 	}
 
 	/// Check if we are stopped.
 	pub fn is_stopped(&self) -> bool {
-		self.stopped
+		self.stopped.load(Ordering::Relaxed)
 	}
 
 	/// Check if we are paused.
 	pub fn is_paused(&self) -> bool {
-		self.paused
+		self.paused.load(Ordering::Relaxed)
 	}
 
 	/// Stop the server.
-	pub fn stop(&mut self) {
+	pub fn stop(&self) {
-		self.stopped = true;
+		self.stopped.store(true, Ordering::Relaxed)
 	}
 
 	/// Pause the server (only used in tests).
-	pub fn pause(&mut self) {
+	pub fn pause(&self) {
-		self.paused = true;
+		self.paused.store(true, Ordering::Relaxed)
 	}
 
 	/// Resume a paused server (only used in tests).
-	pub fn resume(&mut self) {
+	pub fn resume(&self) {
-		self.paused = false;
+		self.paused.store(false, Ordering::Relaxed)
 	}
 }