project-serum
/
solana
mirror of https://github.com/project-serum/solana.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Merge pull request #176 from aeyakovenko/multinode 

Multinode
This commit is contained in:
Greg Fitzgerald 2018-05-07 09:05:12 -06:00 committed by GitHub
parent 6af27669b0 85f83f2c74
commit 880cb8e7cc
8 changed files with 349 additions and 74 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

143
src/accountant_skel.rs
View File

@ -19,6 +19,7 @@ use serde_json;
use signature::PublicKey;
use std::cmp::max;
use std::collections::VecDeque;
use std::io::sink;
use std::io::{Cursor, Write};
use std::mem::size_of;
use std::net::{SocketAddr, UdpSocket};
@ -100,6 +101,8 @@ impl AccountantSkel {
num_events: entry.events.len() as u64,
};
let data = serialize(&Response::EntryInfo(entry_info)).expect("serialize EntryInfo");
trace!("sending {} to {}", data.len(), addr);
//TODO dont do IO here, this needs to be on a separate channel
let res = socket.send_to(&data, addr);
if res.is_err() {
eprintln!("couldn't send response: {:?}", res);
@ -182,16 +185,11 @@ impl AccountantSkel {
broadcast: &streamer::BlobSender,
blob_recycler: &packet::BlobRecycler,
writer: &Arc<Mutex<W>>,
exit: Arc<AtomicBool>,
) -> Result<()> {
let mut q = VecDeque::new();
while let Ok(list) = Self::receive_all(&obj, writer) {
trace!("New blobs? {}", list.len());
Self::process_entry_list_into_blobs(&list, blob_recycler, &mut q);
if exit.load(Ordering::Relaxed) {
break;
}
}
let list = Self::receive_all(&obj, writer)?;
trace!("New blobs? {}", list.len());
Self::process_entry_list_into_blobs(&list, blob_recycler, &mut q);
if !q.is_empty() {
broadcast.send(q)?;
}
@ -206,14 +204,26 @@ impl AccountantSkel {
writer: Arc<Mutex<W>>,
) -> JoinHandle<()> {
spawn(move || loop {
let e = Self::run_sync(
obj.clone(),
&broadcast,
&blob_recycler,
&writer,
exit.clone(),
);
if e.is_err() && exit.load(Ordering::Relaxed) {
let _ = Self::run_sync(obj.clone(), &broadcast, &blob_recycler, &writer);
if exit.load(Ordering::Relaxed) {
info!("sync_service exiting");
break;
}
})
}
/// Process any Entry items that have been published by the Historian.
/// continuosly broadcast blobs of entries out
fn run_sync_no_broadcast(obj: SharedSkel) -> Result<()> {
Self::receive_all(&obj, &Arc::new(Mutex::new(sink())))?;
Ok(())
}
pub fn sync_no_broadcast_service(obj: SharedSkel, exit: Arc<AtomicBool>) -> JoinHandle<()> {
spawn(move || loop {
let _ = Self::run_sync_no_broadcast(obj.clone());
if exit.load(Ordering::Relaxed) {
info!("sync_no_broadcast_service exiting");
break;
}
})
@ -228,7 +238,9 @@ impl AccountantSkel {
match msg {
Request::GetBalance { key } => {
let val = self.acc.lock().unwrap().get_balance(&key);
Some((Response::Balance { key, val }, rsp_addr))
let rsp = (Response::Balance { key, val }, rsp_addr);
info!("Response::Balance {:?}", rsp);
Some(rsp)
}
Request::Transaction(_) => unreachable!(),
Request::Subscribe { subscriptions } => {
@ -247,10 +259,10 @@ impl AccountantSkel {
fn recv_batch(recvr: &streamer::PacketReceiver) -> Result<Vec<SharedPackets>> {
let timer = Duration::new(1, 0);
let msgs = recvr.recv_timeout(timer)?;
trace!("got msgs");
debug!("got msgs");
let mut batch = vec![msgs];
while let Ok(more) = recvr.try_recv() {
trace!("got more msgs");
debug!("got more msgs");
batch.push(more);
}
info!("batch len {}", batch.len());
@ -275,6 +287,7 @@ impl AccountantSkel {
) -> Result<()> {
let batch = Self::recv_batch(recvr)?;
let verified_batches = Self::verify_batch(batch);
debug!("verified batches: {}", verified_batches.len());
for xs in verified_batches {
sendr.send(xs)?;
}
@ -315,8 +328,9 @@ impl AccountantSkel {
&self,
req_vers: Vec<(Request, SocketAddr, u8)>,
) -> Result<Vec<(Response, SocketAddr)>> {
trace!("partitioning");
debug!("partitioning");
let (trs, reqs) = Self::partition_requests(req_vers);
debug!("trs: {} reqs: {}", trs.len(), reqs.len());
// Process the transactions in parallel and then log the successful ones.
for result in self.acc.lock().unwrap().process_verified_transactions(trs) {
@ -328,15 +342,21 @@ impl AccountantSkel {
}
}
debug!("processing verified");
// Let validators know they should not attempt to process additional
// transactions in parallel.
self.historian_input.lock().unwrap().send(Signal::Tick)?;
debug!("after historian_input");
// Process the remaining requests serially.
let rsps = reqs.into_iter()
.filter_map(|(req, rsp_addr)| self.process_request(req, rsp_addr))
.collect();
debug!("returning rsps");
Ok(rsps)
}
@ -377,7 +397,7 @@ impl AccountantSkel {
) -> Result<()> {
let timer = Duration::new(1, 0);
let mms = verified_receiver.recv_timeout(timer)?;
trace!("got some messages: {}", mms.len());
debug!("got some messages: {}", mms.len());
for (msgs, vers) in mms {
let reqs = Self::deserialize_packets(&msgs.read().unwrap());
let req_vers = reqs.into_iter()
@ -389,18 +409,18 @@ impl AccountantSkel {
v
})
.collect();
trace!("process_packets");
debug!("process_packets");
let rsps = obj.process_packets(req_vers)?;
trace!("done process_packets");
debug!("done process_packets");
let blobs = Self::serialize_responses(rsps, blob_recycler)?;
trace!("sending blobs: {}", blobs.len());
if !blobs.is_empty() {
info!("process: sending blobs: {}", blobs.len());
//don't wake up the other side if there is nothing
responder_sender.send(blobs)?;
}
packet_recycler.recycle(msgs);
}
trace!("done responding");
debug!("done responding");
Ok(())
}
/// Process verified blobs, already in order
@ -412,6 +432,7 @@ impl AccountantSkel {
) -> Result<()> {
let timer = Duration::new(1, 0);
let blobs = verified_receiver.recv_timeout(timer)?;
trace!("replicating blobs {}", blobs.len());
for msgs in &blobs {
let blob = msgs.read().unwrap();
let entries: Vec<Entry> = deserialize(&blob.data()[..blob.meta.size]).unwrap();
@ -541,10 +562,12 @@ impl AccountantSkel {
obj: &SharedSkel,
me: ReplicatedData,
gossip: UdpSocket,
serve: UdpSocket,
replicate: UdpSocket,
leader: ReplicatedData,
exit: Arc<AtomicBool>,
) -> Result<Vec<JoinHandle<()>>> {
//replicate pipeline
let crdt = Arc::new(RwLock::new(Crdt::new(me)));
crdt.write().unwrap().set_leader(leader.id);
crdt.write().unwrap().insert(leader);
@ -580,7 +603,7 @@ impl AccountantSkel {
//then sent to the window, which does the erasure coding reconstruction
let t_window = streamer::window(
exit.clone(),
crdt,
crdt.clone(),
blob_recycler.clone(),
blob_receiver,
window_sender,
@ -588,19 +611,76 @@ impl AccountantSkel {
);
let skel = obj.clone();
let t_server = spawn(move || loop {
let s_exit = exit.clone();
let t_replicator = spawn(move || loop {
let e = Self::replicate_state(&skel, &window_receiver, &blob_recycler);
if e.is_err() && exit.load(Ordering::Relaxed) {
if e.is_err() && s_exit.load(Ordering::Relaxed) {
break;
}
});
//serve pipeline
// make sure we are on the same interface
let mut local = serve.local_addr()?;
local.set_port(0);
let respond_socket = UdpSocket::bind(local.clone())?;
let packet_recycler = packet::PacketRecycler::default();
let blob_recycler = packet::BlobRecycler::default();
let (packet_sender, packet_receiver) = channel();
let t_packet_receiver =
streamer::receiver(serve, exit.clone(), packet_recycler.clone(), packet_sender)?;
let (responder_sender, responder_receiver) = channel();
let t_responder = streamer::responder(
respond_socket,
exit.clone(),
blob_recycler.clone(),
responder_receiver,
);
let (verified_sender, verified_receiver) = channel();
let exit_ = exit.clone();
let t_verifier = spawn(move || loop {
let e = Self::verifier(&packet_receiver, &verified_sender);
if e.is_err() && exit_.load(Ordering::Relaxed) {
trace!("verifier exiting");
break;
}
});
let t_sync = Self::sync_no_broadcast_service(obj.clone(), exit.clone());
let skel = obj.clone();
let s_exit = exit.clone();
let t_server = spawn(move || loop {
let e = Self::process(
&mut skel.clone(),
&verified_receiver,
&responder_sender,
&packet_recycler,
&blob_recycler,
);
if e.is_err() {
if s_exit.load(Ordering::Relaxed) {
break;
}
}
});
Ok(vec![
//replicate threads
t_blob_receiver,
t_retransmit,
t_window,
t_server,
t_replicator,
t_gossip,
t_listen,
//serve threads
t_packet_receiver,
t_responder,
t_server,
t_verifier,
t_sync,
])
}
}
@ -769,7 +849,7 @@ mod tests {
tr2.data.plan = Plan::new_payment(502, bob_pubkey);
let _sig = acc_stub.transfer_signed(tr2).unwrap();
assert_eq!(acc_stub.get_balance(&bob_pubkey).wait().unwrap(), 500);
assert_eq!(acc_stub.get_balance(&bob_pubkey).unwrap(), 500);
trace!("exiting");
exit.store(true, Ordering::Relaxed);
trace!("joining threads");
@ -797,7 +877,7 @@ mod tests {
fn test_replicate() {
logger::setup();
let (leader_data, leader_gossip, _, leader_serve) = test_node();
let (target1_data, target1_gossip, target1_replicate, _) = test_node();
let (target1_data, target1_gossip, target1_replicate, target1_serve) = test_node();
let (target2_data, target2_gossip, target2_replicate, _) = test_node();
let exit = Arc::new(AtomicBool::new(false));
@ -851,6 +931,7 @@ mod tests {
&acc,
target1_data,
target1_gossip,
target1_serve,
target1_replicate,
leader_data,
exit.clone(),

160
src/accountant_stub.rs
View File

@ -47,7 +47,9 @@ impl AccountantStub {
pub fn recv_response(&self) -> io::Result<Response> {
let mut buf = vec![0u8; 1024];
info!("start recv_from");
self.socket.recv_from(&mut buf)?;
info!("end recv_from");
let resp = deserialize(&buf).expect("deserialize balance");
Ok(resp)
}
@ -55,9 +57,11 @@ impl AccountantStub {
pub fn process_response(&mut self, resp: Response) {
match resp {
Response::Balance { key, val } => {
info!("Response balance {:?} {:?}", key, val);
self.balances.insert(key, val);
}
Response::EntryInfo(entry_info) => {
trace!("Response entry_info {:?}", entry_info.id);
self.last_id = Some(entry_info.id);
self.num_events += entry_info.num_events;
}
@ -88,7 +92,8 @@ impl AccountantStub {
/// Request the balance of the user holding `pubkey`. This method blocks
/// until the server sends a response. If the response packet is dropped
/// by the network, this method will hang indefinitely.
pub fn get_balance(&mut self, pubkey: &PublicKey) -> FutureResult<i64, i64> {
pub fn get_balance(&mut self, pubkey: &PublicKey) -> io::Result<i64> {
info!("get_balance");
let req = Request::GetBalance { key: *pubkey };
let data = serialize(&req).expect("serialize GetBalance");
self.socket
@ -96,13 +101,14 @@ impl AccountantStub {
.expect("buffer error");
let mut done = false;
while !done {
let resp = self.recv_response().expect("recv response");
let resp = self.recv_response()?;
info!("recv_response {:?}", resp);
if let &Response::Balance { ref key, .. } = &resp {
done = key == pubkey;
}
self.process_response(resp);
}
ok(self.balances[pubkey].unwrap())
self.balances[pubkey].ok_or(io::Error::new(io::ErrorKind::Other, "nokey"))
}
/// Request the last Entry ID from the server. This method blocks
@ -143,21 +149,23 @@ mod tests {
use super::*;
use accountant::Accountant;
use accountant_skel::AccountantSkel;
use crdt::ReplicatedData;
use crdt::{Crdt, ReplicatedData};
use futures::Future;
use historian::Historian;
use logger;
use mint::Mint;
use signature::{KeyPair, KeyPairUtil};
use std::io::sink;
use std::sync::atomic::{AtomicBool, Ordering};
use std::sync::mpsc::sync_channel;
use std::sync::Arc;
use std::sync::{Arc, RwLock};
use std::thread::sleep;
use std::time::Duration;
// TODO: Figure out why this test sometimes hangs on TravisCI.
#[test]
fn test_accountant_stub() {
logger::setup();
let gossip = UdpSocket::bind("0.0.0.0:0").unwrap();
let serve = UdpSocket::bind("0.0.0.0:0").unwrap();
let addr = serve.local_addr().unwrap();
@ -180,16 +188,154 @@ mod tests {
sleep(Duration::from_millis(300));
let socket = UdpSocket::bind("0.0.0.0:0").unwrap();
socket.set_read_timeout(Some(Duration::new(5, 0))).unwrap();
let mut acc = AccountantStub::new(addr, socket);
let last_id = acc.get_last_id().wait().unwrap();
let _sig = acc.transfer(500, &alice.keypair(), bob_pubkey, &last_id)
.unwrap();
assert_eq!(acc.get_balance(&bob_pubkey).wait().unwrap(), 500);
assert_eq!(acc.get_balance(&bob_pubkey).unwrap(), 500);
exit.store(true, Ordering::Relaxed);
for t in threads {
t.join().unwrap();
}
}
fn test_node() -> (ReplicatedData, UdpSocket, UdpSocket, UdpSocket) {
let gossip = UdpSocket::bind("0.0.0.0:0").unwrap();
let serve = UdpSocket::bind("0.0.0.0:0").unwrap();
let replicate = UdpSocket::bind("0.0.0.0:0").unwrap();
let pubkey = KeyPair::new().pubkey();
let leader = ReplicatedData::new(
pubkey,
gossip.local_addr().unwrap(),
replicate.local_addr().unwrap(),
serve.local_addr().unwrap(),
);
(leader, gossip, serve, replicate)
}
#[test]
fn test_multi_accountant_stub() {
logger::setup();
info!("test_multi_accountant_stub");
let leader = test_node();
let replicant = test_node();
let alice = Mint::new(10_000);
let bob_pubkey = KeyPair::new().pubkey();
let exit = Arc::new(AtomicBool::new(false));
let leader_acc = {
let (input, event_receiver) = sync_channel(10);
let historian = Historian::new(event_receiver, &alice.last_id(), Some(30));
let acc = Accountant::new(&alice);
Arc::new(AccountantSkel::new(acc, input, historian))
};
let replicant_acc = {
let (input, event_receiver) = sync_channel(10);
let historian = Historian::new(event_receiver, &alice.last_id(), Some(30));
let acc = Accountant::new(&alice);
Arc::new(AccountantSkel::new(acc, input, historian))
};
let leader_threads = AccountantSkel::serve(
&leader_acc,
leader.0.clone(),
leader.2,
leader.1,
exit.clone(),
sink(),
).unwrap();
let replicant_threads = AccountantSkel::replicate(
&replicant_acc,
replicant.0.clone(),
replicant.1,
replicant.2,
replicant.3,
leader.0.clone(),
exit.clone(),
).unwrap();
//lets spy on the network
let (mut spy, spy_gossip, _, _) = test_node();
let daddr = "0.0.0.0:0".parse().unwrap();
spy.replicate_addr = daddr;
spy.serve_addr = daddr;
let mut spy_crdt = Crdt::new(spy);
spy_crdt.insert(leader.0.clone());
spy_crdt.set_leader(leader.0.id);
let spy_ref = Arc::new(RwLock::new(spy_crdt));
let t_spy_listen = Crdt::listen(spy_ref.clone(), spy_gossip, exit.clone());
let t_spy_gossip = Crdt::gossip(spy_ref.clone(), exit.clone());
//wait for the network to converge
for _ in 0..20 {
let ix = spy_ref.read().unwrap().update_index;
info!("my update index is {}", ix);
let len = spy_ref.read().unwrap().remote.values().len();
let mut done = false;
info!("remote len {}", len);
if len > 1 && ix > 2 {
done = true;
//check if everyones remote index is greater or equal to ours
let vs: Vec<u64> = spy_ref.read().unwrap().remote.values().cloned().collect();
for t in vs.into_iter() {
info!("remote update index is {} vs {}", t, ix);
if t < 3 {
done = false;
}
}
}
if done == true {
info!("converged!");
break;
}
sleep(Duration::new(1, 0));
}
//verify leader can do transfer
let leader_balance = {
let socket = UdpSocket::bind("0.0.0.0:0").unwrap();
socket.set_read_timeout(Some(Duration::new(1, 0))).unwrap();
let mut acc = AccountantStub::new(leader.0.serve_addr, socket);
info!("getting leader last_id");
let last_id = acc.get_last_id().wait().unwrap();
info!("executing leader transer");
let _sig = acc.transfer(500, &alice.keypair(), bob_pubkey, &last_id)
.unwrap();
info!("getting leader balance");
acc.get_balance(&bob_pubkey).unwrap()
};
assert_eq!(leader_balance, 500);
//verify replicant has the same balance
let mut replicant_balance = 0;
for _ in 0..10 {
let socket = UdpSocket::bind("0.0.0.0:0").unwrap();
socket.set_read_timeout(Some(Duration::new(1, 0))).unwrap();
let mut acc = AccountantStub::new(replicant.0.serve_addr, socket);
info!("getting replicant balance");
if let Ok(bal) = acc.get_balance(&bob_pubkey) {
replicant_balance = bal;
}
info!("replicant balance {}", replicant_balance);
if replicant_balance == leader_balance {
break;
}
sleep(Duration::new(1, 0));
}
assert_eq!(replicant_balance, leader_balance);
exit.store(true, Ordering::Relaxed);
for t in leader_threads {
t.join().unwrap();
}
for t in replicant_threads {
t.join().unwrap();
}
for t in vec![t_spy_listen, t_spy_gossip] {
t.join().unwrap();
}
}
}

28
src/bin/client-demo.rs
View File

@ -18,6 +18,8 @@ use std::env;
use std::io::{stdin, Read};
use std::net::{SocketAddr, UdpSocket};
use std::process::exit;
use std::thread::sleep;
use std::time::Duration;
use std::time::Instant;
use untrusted::Input;
@ -38,7 +40,7 @@ fn main() {
let mut opts = Options::new();
opts.optopt("s", "", "server address", "host:port");
opts.optopt("c", "", "client address", "host:port");
opts.optopt("t", "", "number of threads", "4");
opts.optopt("t", "", "number of threads", &format!("{}", threads));
opts.optflag("h", "help", "print help");
let args: Vec<String> = env::args().collect();
let matches = match opts.parse(&args[1..]) {
@ -84,6 +86,7 @@ fn main() {
println!("Binding to {}", client_addr);
let socket = UdpSocket::bind(&client_addr).unwrap();
socket.set_read_timeout(Some(Duration::new(5, 0))).unwrap();
let mut acc = AccountantStub::new(addr.parse().unwrap(), socket);
println!("Get last ID...");
@ -104,7 +107,7 @@ fn main() {
.into_par_iter()
.map(|chunk| Transaction::new(&chunk[0], chunk[1].pubkey(), 1, last_id))
.collect();
let duration = now.elapsed();
let mut duration = now.elapsed();
let ns = duration.as_secs() * 1_000_000_000 + u64::from(duration.subsec_nanos());
let bsps = txs as f64 / ns as f64;
let nsps = ns as f64 / txs as f64;
@ -115,6 +118,7 @@ fn main() {
);
let initial_tx_count = acc.transaction_count();
println!("initial count {}", initial_tx_count);
println!("Transfering {} transactions in {} batches", txs, threads);
let now = Instant::now();
@ -131,16 +135,16 @@ fn main() {
}
});
println!("Waiting for half the transactions to complete...",);
let mut tx_count = acc.transaction_count();
while tx_count < transactions.len() as u64 / 2 {
println!("Waiting for transactions to complete...",);
let mut tx_count;
for _ in 0..5 {
tx_count = acc.transaction_count();
duration = now.elapsed();
let txs = tx_count - initial_tx_count;
println!("Transactions processed {}", txs);
let ns = duration.as_secs() * 1_000_000_000 + u64::from(duration.subsec_nanos());
let tps = (txs * 1_000_000_000) as f64 / ns as f64;
println!("{} tps", tps);
sleep(Duration::new(1, 0));
}
let txs = tx_count - initial_tx_count;
println!("Transactions processed {}", txs);
let duration = now.elapsed();
let ns = duration.as_secs() * 1_000_000_000 + u64::from(duration.subsec_nanos());
let tps = (txs * 1_000_000_000) as f64 / ns as f64;
println!("Done. {} tps", tps);
}

9
src/bin/testnode.rs
View File

@ -76,6 +76,8 @@ fn main() {
})
});
eprintln!("done parsing...");
// The first item in the ledger is required to be an entry with zero num_hashes,
// which implies its id can be used as the ledger's seed.
let entry0 = entries.next().unwrap();
@ -90,10 +92,14 @@ fn main() {
None
};
eprintln!("creating accountant...");
let acc = Accountant::new_from_deposit(&deposit.unwrap());
acc.register_entry_id(&entry0.id);
acc.register_entry_id(&entry1.id);
eprintln!("processing entries...");
let mut last_id = entry1.id;
for entry in entries {
last_id = entry.id;
@ -101,6 +107,8 @@ fn main() {
acc.register_entry_id(&last_id);
}
eprintln!("creating networking stack...");
let (input, event_receiver) = sync_channel(10_000);
let historian = Historian::new(event_receiver, &last_id, Some(1000));
let exit = Arc::new(AtomicBool::new(false));
@ -115,6 +123,7 @@ fn main() {
replicate_sock.local_addr().unwrap(),
serve_sock.local_addr().unwrap(),
);
eprintln!("starting server...");
let threads =
AccountantSkel::serve(&skel, d, serve_sock, gossip_sock, exit.clone(), stdout()).unwrap();
eprintln!("Ready. Listening on {}", serve_addr);

69
src/crdt.rs
View File

@ -91,7 +91,7 @@ pub struct Crdt {
local: HashMap<PublicKey, u64>,
/// The value of the remote update index that i have last seen
/// This Node will ask external nodes for updates since the value in this list
remote: HashMap<PublicKey, u64>,
pub remote: HashMap<PublicKey, u64>,
pub update_index: u64,
me: PublicKey,
timeout: Duration,
@ -172,20 +172,33 @@ impl Crdt {
let cloned_table: Vec<ReplicatedData> = robj.table.values().cloned().collect();
(robj.table[&robj.me].clone(), cloned_table)
};
let errs: Vec<_> = table
let daddr = "0.0.0.0:0".parse().unwrap();
let items: Vec<(usize, &ReplicatedData)> = table
.iter()
.enumerate()
.cycle()
.zip(blobs.iter())
.map(|((i, v), b)| {
.filter(|v| {
if me.id == v.id {
return Ok(0);
//filter myself
false
} else if v.replicate_addr == daddr {
//filter nodes that are not listening
false
} else {
true
}
})
.enumerate()
.collect();
let orders: Vec<_> = items.into_iter().cycle().zip(blobs.iter()).collect();
let errs: Vec<_> = orders
.into_par_iter()
.map(|((i, v), b)| {
// only leader should be broadcasting
assert!(me.current_leader_id != v.id);
let mut blob = b.write().unwrap();
blob.set_id(me.id).expect("set_id");
blob.set_index(*transmit_index + i as u64)
.expect("set_index");
//TODO profile this, may need multiple sockets for par_iter
s.send_to(&blob.data[..blob.meta.size], &v.replicate_addr)
})
.collect();
@ -210,17 +223,28 @@ impl Crdt {
(s.table[&s.me].clone(), s.table.values().cloned().collect())
};
let rblob = blob.read().unwrap();
let errs: Vec<_> = table
let daddr = "0.0.0.0:0".parse().unwrap();
let orders: Vec<_> = table
.iter()
.filter(|v| {
if me.id == v.id {
false
} else if me.current_leader_id == v.id {
trace!("skip retransmit to leader {:?}", v.id);
false
} else if v.replicate_addr == daddr {
trace!("skip nodes that are not listening {:?}", v.id);
false
} else {
true
}
})
.collect();
let errs: Vec<_> = orders
.par_iter()
.map(|v| {
if me.id == v.id {
return Ok(0);
}
if me.current_leader_id == v.id {
trace!("skip retransmit to leader{:?}", v.id);
return Ok(0);
}
trace!("retransmit blob to {}", v.replicate_addr);
//TODO profile this, may need multiple sockets for par_iter
s.send_to(&rblob.data[..rblob.meta.size], &v.replicate_addr)
})
.collect();
@ -258,13 +282,18 @@ impl Crdt {
/// (A,B)
/// * A - Address to send to
/// * B - RequestUpdates protocol message
fn gossip_request(&self) -> (SocketAddr, Protocol) {
let n = (Self::random() as usize) % self.table.len();
trace!("random {:?} {}", &self.me[0..1], n);
fn gossip_request(&self) -> Result<(SocketAddr, Protocol)> {
if self.table.len() <= 1 {
return Err(Error::GeneralError);
}
let mut n = (Self::random() as usize) % self.table.len();
while self.table.values().nth(n).unwrap().id == self.me {
n = (Self::random() as usize) % self.table.len();
}
let v = self.table.values().nth(n).unwrap().clone();
let remote_update_index = *self.remote.get(&v.id).unwrap_or(&0);
let req = Protocol::RequestUpdates(remote_update_index, self.table[&self.me].clone());
(v.gossip_addr, req)
Ok((v.gossip_addr, req))
}
/// At random pick a node and try to get updated changes from them
@ -274,7 +303,7 @@ impl Crdt {
// Lock the object only to do this operation and not for any longer
// especially not when doing the `sock.send_to`
let (remote_gossip_addr, req) = obj.read().unwrap().gossip_request();
let (remote_gossip_addr, req) = obj.read().unwrap().gossip_request()?;
let sock = UdpSocket::bind("0.0.0.0:0")?;
// TODO this will get chatty, so we need to first ask for number of updates since
// then only ask for specific data that we dont have

8
src/packet.rs
View File

@ -17,6 +17,7 @@ pub type BlobRecycler = Recycler<Blob>;
pub const NUM_PACKETS: usize = 1024 * 8;
pub const BLOB_SIZE: usize = 64 * 1024;
pub const BLOB_DATA_SIZE: usize = BLOB_SIZE - BLOB_ID_END;
pub const PACKET_DATA_SIZE: usize = 256;
pub const NUM_BLOBS: usize = (NUM_PACKETS * PACKET_DATA_SIZE) / BLOB_SIZE;
@ -178,13 +179,14 @@ impl Packets {
socket.set_nonblocking(false)?;
for p in &mut self.packets {
p.meta.size = 0;
trace!("receiving");
match socket.recv_from(&mut p.data) {
Err(_) if i > 0 => {
trace!("got {:?} messages", i);
debug!("got {:?} messages", i);
break;
}
Err(e) => {
info!("recv_from err {:?}", e);
trace!("recv_from err {:?}", e);
return Err(Error::IO(e));
}
Ok((nrecv, from)) => {
@ -202,6 +204,7 @@ impl Packets {
pub fn recv_from(&mut self, socket: &UdpSocket) -> Result<()> {
let sz = self.run_read_from(socket)?;
self.packets.resize(sz, Packet::default());
debug!("recv_from: {}", sz);
Ok(())
}
pub fn send_to(&self, socket: &UdpSocket) -> Result<()> {
@ -233,6 +236,7 @@ impl Blob {
let e = deserialize(&self.data[BLOB_INDEX_END..BLOB_ID_END])?;
Ok(e)
}
pub fn set_id(&mut self, id: PublicKey) -> Result<()> {
let wtr = serialize(&id)?;
self.data[BLOB_INDEX_END..BLOB_ID_END].clone_from_slice(&wtr);

5
src/recorder.rs
View File

@ -8,6 +8,7 @@
use entry::{create_entry_mut, Entry};
use event::Event;
use hash::{hash, Hash};
use packet::BLOB_DATA_SIZE;
use std::mem;
use std::sync::mpsc::{Receiver, SyncSender, TryRecvError};
use std::time::{Duration, Instant};
@ -83,7 +84,7 @@ impl Recorder {
// Record an entry early if we anticipate its serialized size will
// be larger than 64kb. At the time of this writing, we assume each
// event will be well under 256 bytes.
if self.events.len() >= 65_536 / 256 {
if self.events.len() >= BLOB_DATA_SIZE / 256 {
self.record_entry()?;
}
}
@ -100,8 +101,8 @@ mod tests {
use super::*;
use bincode::serialize;
use signature::{KeyPair, KeyPairUtil};
use transaction::Transaction;
use std::sync::mpsc::sync_channel;
use transaction::Transaction;
#[test]
fn test_sub64k_entry_size() {

1
src/result.rs
View File

@ -18,6 +18,7 @@ pub enum Error {
AccountingError(accountant::AccountingError),
SendError,
Services,
GeneralError,
}
pub type Result<T> = std::result::Result<T, Error>;