Why is UdpSocket::recv_from stuck forever? - rust

I have the following client IPMI ping program:
pub fn ping(host: &String) -> Result<(), String> {
let socket = UdpSocket::bind("0.0.0.0:0").expect("Unable to bind to host");
let mut udp_payload: Vec<u8> = vec![];
// WRITE RMCP header
// WRITE ASF Ping header
let mut retry_count = 3;
while retry_count > 0 {
match socket.send_to(udp_payload.as_slice(), host) {
Ok(_) => trace!("Successfully sent ping packet"),
Err(_) => {
trace!("Unable to send ping packet retrying");
retry_count -= 1;
continue;
}
}
trace!("Sent ping pkt");
let mut buffer = [0u8; 1500];
match socket.recv_from(&mut buffer) {
Ok((amt, source)) => {
trace!("Got {} a pong response from {}", amt, source);
// Verify RMCP and ASF Pong headers
return Ok(());
}
Err(err) => {
trace!("Did not receive a Pong response, so retrying {}", err);
retry_count = retry_count - 1;
continue;
}
}
}
Err(format!("IPMI ping failed"))
}
I took a packet capture while running the above function. I see the ping packet being sent and a pong response being received from the server, but the recv_from function never returns.

I was running the above program on Mac along with Proxifier that was messing up the UDP traffic. Once I stopped running proxifier my program start working without any issues.

I don't know....
But I would guess that "Proxifier" changes the firewall settings, and those changes make disrupt your UDP. You might want to read up on iptables, and its replacement nftables.

Related

Rust TcpStream doesn't send data to TcpListener until program is stopped

I'm making a simple chat using Tokio for learning. I've made a server with TcpListener that works fine when I use telnets conexions. For example, if I open 2 terminals with telnet and send data, the server receives the conexion and send the data to the address that is not the sender.
The problem I have is with the TcpStream. For some reason (I think is some type of blocking problem) the stream is not sending the data until I stop the app.
I'd tried multiple configurations, but I think that the simple requiered that is close to "work" is the next one.
#[tokio::main]
async fn main() -> Result<(), Box<dyn Error>> {
let mut stream = TcpStream::connect("127.0.0.1:8080").await?;
println!("INFO: Socket Address: {:?}", stream.local_addr().unwrap());
let (read, mut writer) = stream.split();
let (sender, receiver) = channel();
thread::spawn(move || loop {
let mut line_buf = String::new();
stdin().read_line(&mut line_buf);
let line = line_buf.trim_end().to_string();
if line.len() > 0 {
sender.send(line_buf.trim_end().to_string()).unwrap();
}
});
loop {
let line = receiver.recv().unwrap();
println!("DEBUG User Input {line}");
writer.write_all(line.as_bytes()).await?;
}
P.D: this code doesn't have anything for receive the messages sended by the server. For manage that, I had something like:
tokio::spawn(async move {
let (read, mut writer) = socket.split();
let mut reader = BufReader::new(read);
let mut line = String::new();
loop {
tokio::select! {
result = reader.read_line(&mut line) => {
if result.unwrap() == 0 {
break;
}
print!("DEBUG: Message Received: {line}");
tx.send((line.clone(), addr)).unwrap();
line.clear();
}
result = rx.recv() => {
let (msg, other_addr) = result.unwrap();
println!("INFO: Message Address: {addr}\n\n");
if addr != other_addr {
writer.write_all(msg.as_bytes()).await.unwrap();
}
}
}
}
});

How do I drop the connection in my simple tcp server?

I took a really simple example of a rust server with tcp.
use std::net::{Shutdown,TcpListener, TcpStream};
use std::thread;
use std::io::{Read,Write,Error};
fn handle_client(mut stream: TcpStream)-> Result<(), Error> {
println!("incoming connection from: {}", stream.peer_addr()?);
let mut buf = [0;512];
loop {
let bytes_read = stream.read(&mut buf)?;
if bytes_read == 0 {return Ok(())}
let tmp = format!("{}", String::from_utf8_lossy(&buf).trim());
eprintln!("getting {}",tmp);
stream.write(&buf[..bytes_read])?;
}
}
fn main() {
let listener = TcpListener::bind("0.0.0.0:8888").expect("Could not bind");
for stream in listener.incoming() {
match stream {
Err(e)=> {eprintln!("failed: {}", e)}
Ok(stream) => {
thread::spawn(move || {
handle_client(stream).unwrap_or_else(|error| eprintln!("{:?}", error));
});
}
}
}
}
Which basically takes input, spits it back at the client, and prints to it's own terminal.
I would like to be able to end this connection. Ending the connection should probably happen depending on something, but right now I want to just try to shut it down.
I tried looking thorugh the docs, and then tried adding the shutdown method
Now, I want to take some stream as input, do something with it, and then shut the channel
So I tried doing this:
fn main() {
let listener = TcpListener::bind("0.0.0.0:8888").expect("Could not bind");
for stream in listener.incoming() {
match stream {
Err(e)=> {eprintln!("failed: {}", e)}
Ok(stream) => {
thread::spawn(move || {
handle_client(stream).unwrap_or_else(|error| eprintln!("{:?}", error));
});
stream.shutdown(Shutdown::Both).expect("shutdown call failed");
}
}
}
}
But this causes an issue with the "stream" being a moved value.
So how can I shut down the channel right after receiving and doing something with the input?
(I still want to preserve this structure with the loop, since I actually want to receive many messages, and then shut down dependning on the input)
You can try to do a clone of the stream and pass that stream clone to the function you spawn. Then you can call the shutdown inside your spawned function after you have handled the client.
This way, your original stream variable would remain intact and you can have the behavior you want.
The reason you get this issue is that the spawned function moves everything it captures due to the move keyword.
fn main() {
let listener = TcpListener::bind("0.0.0.0:8888").expect("Could not bind");
for stream in listener.incoming() {
let stream_clone = stream.clone();
match stream {
Err(e)=> {eprintln!("failed: {}", e)}
Ok(stream) => {
thread::spawn(move || {
handle_client(stream_clone).unwrap_or_else(|error| eprintln!("{:?}", error));
stream_clone.shutdown(Shutdown::Both).expect("shutdown call failed");
});
}
}
}
}
Also, you would need to change your signature of handle client to fn handle_client(stream: &mut TcpStream)-> Result<(), Error> so it would not also move the cloned stream variable, but burrow as mutable instead.

Receiver on "tokio::mpsc::channel" only receives messages when buffer is full

In my code snippet the tokio (v0.3) mpsc:channel receiver only receives a message when the buffer is full. It doesn't matter how big or small the buffer is.
use std::io;
use std::net::{SocketAddr, ToSocketAddrs};
use std::sync::Arc;
use std::time::Duration;
use tokio::net::UdpSocket;
use tokio::sync::mpsc;
use tokio::time::sleep;
const MESSAGE_LENGTH: usize = 1024;
pub struct Peer {
socket: Arc<UdpSocket>,
}
impl Peer {
pub fn new<S: ToSocketAddrs>(addr: S) -> Peer {
let socket = std::net::UdpSocket::bind(addr).expect("could not create socket");
let peer = Peer {
socket: Arc::new(UdpSocket::from_std(socket).unwrap()),
};
peer.start_inbound_message_handler();
peer
}
pub fn local_addr(&self) -> SocketAddr {
self.socket.local_addr().unwrap()
}
fn start_inbound_message_handler(&self) {
let socket = self.socket.clone();
let (tx, rx) = mpsc::channel(1);
self.start_request_handler(rx);
tokio::spawn(async move {
let mut buf = [0u8; MESSAGE_LENGTH];
loop {
if let Ok((len, addr)) = socket.recv_from(&mut buf).await {
println!("received {} bytes from {}", len, addr);
if let Err(_) = tx.send(true).await {
println!("error sending msg to request handler");
}
}
}
});
}
fn start_request_handler(&self, mut receiver: mpsc::Receiver<bool>) {
tokio::spawn(async move {
while let Some(msg) = receiver.recv().await {
println!("got ping request: {:?}", msg);
}
});
}
pub async fn send_ping(&self, dest: String) -> Result<(), io::Error> {
let buf = [255u8; MESSAGE_LENGTH];
self.socket.send_to(&buf[..], &dest).await?;
Ok(())
}
}
#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
let peer1 = Peer::new("0.0.0.0:0");
println!("peer1 started on: {}", peer1.local_addr().to_string());
let peer2 = Peer::new("0.0.0.0:0");
println!("peer2 started on: {}", peer2.local_addr().to_string());
peer2.send_ping(peer1.local_addr().to_string()).await?;
peer2.send_ping(peer1.local_addr().to_string()).await?;
sleep(Duration::from_secs(100)).await;
Ok(())
}
Link to the Playground
In the start_inbound_message_handler function I start reading from the socket, if a message was received, a message over the mpsc::channel would be send to the start_request_handler where the processing happens, in this case a simple log output would be written if the receiver receives anything.
In the main function I'm creating two peers, peer1 and peer2, after both peers are created I start a ping request to the first peer. In the start_inbound_message_handler I will receive the data from the udp socket and send a message over the mpsc::channel The send returns without error. The problem is as mentioned before that the receiver will only receive a message when the buffer is full. In this case the buffer is 1. So if I send a second ping the first ping is received. I cannot find out why this happen.
The expected behavior is, if I send a message over the channel, the receiver starts receiving messages immediately and is not waiting until the buffer is full.
According to the Tokio documentation of from_std():
Creates new UdpSocket from a previously bound std::net::UdpSocket.
This function is intended to be used to wrap a UDP socket from the
standard library in the Tokio equivalent. The conversion assumes nothing
about the underlying socket; it is left up to the user to set it in
non-blocking mode.
This can be used in conjunction with socket2's Socket interface to
configure a socket before it's handed off, such as setting options like
reuse_address or binding to multiple addresses.
A socket that is not in non-blocking mode will prevent Tokio from working normally.
Just use the tokio function bind(), it is way simpler.

Rust TCP socket server only working with one connection

I'm new to Rust and I'm trying to configure a simple tcp socket server which will listen to connections and will reply with the same message that received.
The thing is, this works as I want except when connecting with multiple clients.. The first client that connects will send and receive the messages but if a second client connects, the first one keeps working but the second never receives messages, in fact the message never gets in the code that will handle it. And if I disconnect the first socket, the server will start spamming forever that received a message from the first socket with the same content than the last message it sent.
I am pretty sure I did something wrong in my code but I can't find it
This is my server struct:
use std::collections::HashMap;
use std::io::Read;
use std::io::Write;
use std::net::Shutdown;
use std::net::TcpListener;
use std::net::TcpStream;
use std::str;
use std::sync::{Arc, RwLock};
use threadpool::ThreadPool;
#[derive(Clone, Debug)]
pub struct Server {
id: Arc<RwLock<u32>>,
connections: Arc<RwLock<HashMap<u32, TcpStream>>>,
url: String,
thread_pool: ThreadPool
}
impl Server {
pub fn new(url: String) -> Server {
let server = Server {
id: Arc::new(RwLock::new(0)),
connections: Arc::new(RwLock::new(HashMap::new())),
url,
thread_pool: ThreadPool::new(10)
};
server
}
pub fn start(&self) {
let listener = TcpListener::bind(&self.url).expect("Could not start the server");
println!("Server started succesfully");
for stream in listener.incoming() {
match stream {
Ok(stream) => {
let mut self_clone = self.clone();
self.thread_pool.execute(move || {
self_clone.on_client_connect(stream.try_clone().unwrap());
});
}
Err(error) => eprintln!("Error when tried to use stream. Error = {:?}", error),
}
}
}
fn on_client_connect(&mut self, stream: TcpStream) {
println!("Client connected from {}", stream.local_addr().unwrap());
let mut id = self.id.write().unwrap();
{
*id += 1;
}
self.connections
.write()
.unwrap()
.insert(*id, stream.try_clone().unwrap());
let mut stream = stream.try_clone().unwrap();
let mut buffer = [0; 1024];
while match stream.read(&mut buffer) {
Ok(size) => {
println!(
"Message received from {} - {}",
id,
str::from_utf8(&buffer).unwrap()
);
stream.write_all(&buffer[0..size]).unwrap();
true
}
Err(error) => {
println!(
"Error when reading message from socket. Error = {:?}",
error
);
stream.shutdown(Shutdown::Both).unwrap();
false
}
} { }
}
}
And in my main.rs I'm just calling the connect function and the server starts working
In this piece of code in your on_client_connect function, you're aquiring a read lock for self.id:
let mut id = self.id.write().unwrap();
{
*id += 1;
}
However, the id variable, which holds the lock, is not released until it drops at the end of the function. This means that all other clients will wait for this lock to be released, which won't happen until the function currently holding the lock has completed (which happens when that client disconnects).
You can solve this by rewriting the above code to only keep the lock while incrementing, and then storing the ID value in a variable:
let id: u32 = {
let mut id_lock = self.id.write.unwrap();
*id_lock += 1;
*id_lock
// id_lock is dropped at the end of this block, so the lock is released
};
Even better, you can use AtomicU32, which is still thread-safe yet does not require locking at all:
use std::sync::atomic::{AtomicU32, Ordering};
struct {
id: Arc<AtomicU32>,
// ...
}
// Fetch previous value, then increment `self.id` by one, in a thread-safe and lock-free manner
let id: u32 = self.id.fetch_add(1, Ordering::Relaxed);
Also, when the connection is closed your code goes into an infinite loop because you're not handling the case where stream.read() returns Ok(0), which indicates that the connection was closed:
while match stream.read(&mut buffer) {
Ok(0) => false, // handle connection closed...
Ok(size) => { /* ... */ }
Err(err) => { /* ... */ }
} {}

Tokio future that reads from a channel and uses poll_fn and try_ready never completes

I have a Tokio future which never completes (rx is a Receiver and sock is a tokio UdpSocket). It basically reads packets from a packet queue and transmits them over a socket:
poll_fn(move || {
match try_ready!(rx
.poll()
.map_err(|_e| tokio::io::Error::new(tokio::io::ErrorKind::Other, "Poll error")))
{
Some((packet, to)) => {
println!(
"Rx: Received {} bytes for {}: {:?}",
packet.len(),
to,
packet.as_slice(),
);
try_ready!(sock.poll_send_to(packet.as_slice(), &to));
println!("Sent");
}
None => println!("Rx end"),
}
Ok(futures::Async::Ready(()))
})
.map_err(|e: tokio::io::Error| println!("Error: {:?}", e))
It executes until the poll_send_to line (the println! just before poll_send_to executes, the println! just after doesn't) and then waits forever without sending the packet.
I replaced the above future with the following one to ensure that it wasn't a socket issue(had some issues with what I think were flaky notifications before):
poll_fn(move || {
let packet = vec![0;10];
let to = SocketAddr::from_str("127.0.0.1:8001").expect("Parse error");
try_ready!(sock.poll_send_to(packet.as_slice(), &to));
Ok(futures::Async::Ready(()))
})
.map_err(|e: tokio::io::Error| println!("Error: {:?}", e))
This future worked perfectly - it sent the packet as expected and exited the program.
I don't think the problem is with the message channels given that rx can poll successfully and prints the println message. I don't think the problem is with the socket either given that the second future works. I am observing packets directly through Wireshark, so I don't think it is an issue with my observations either.
I'm pretty new to Rust and Tokio, so it's possible that I am overlooking some basic fact (e.g. can't try_ready twice in the same future, future doesn't resume from where it left off previously, etc).
Can you help me figure out the problem with the first future?
use futures::future::lazy;
use futures::stream::Stream;
use futures::try_ready;
use std::net::SocketAddr;
use std::str::FromStr;
use tokio;
use tokio::net::UdpSocket;
use tokio::prelude::future::poll_fn;
use tokio::prelude::Future;
fn main() {
let mut sock = UdpSocket::bind(&SocketAddr::from_str("127.0.0.1:8000").expect("Parse error"))
.expect("Bind error");
let (mut tx, mut rx) = tokio::sync::mpsc::channel::<(Vec<u8>, SocketAddr)>(2000);
tokio::run(lazy(move || {
//----------------- This future works ----------------//
// tokio::spawn(
// poll_fn(move || {
// let packet = vec![70; 10];
// let to = SocketAddr::from_str("127.0.0.1:8001").expect("Parse error");
// try_ready!(sock.poll_send_to(packet.as_slice(), &to));
// Ok(futures::Async::Ready(()))
// })
// .map_err(|e: tokio::io::Error| println!("Error: {:?}", e)),
// );
//----------------- This future doesn't ----------------//
tokio::spawn(
poll_fn(move || {
match try_ready!(rx
.poll()
.map_err(|_e| tokio::io::Error::new(tokio::io::ErrorKind::Other, "Poll error")))
{
Some((packet, to)) => {
// This is printed
println!(
"Rx: Received {} bytes for {}: {:?}",
packet.len(),
to,
packet.as_slice(),
);
try_ready!(sock.poll_send_to(packet.as_slice(), &to));
// This is never printed
println!("Sent");
}
None => println!("Rx end"),
}
Ok(futures::Async::Ready(()))
})
.map_err(|e: tokio::io::Error| println!("Error: {:?}", e)),
);
//----------------- This future queues a packet ----------------//
tokio::spawn(
poll_fn(move || {
try_ready!(tx.poll_ready());
tx.try_send((
vec![70; 10],
SocketAddr::from_str("127.0.0.1:8001").expect("Parse error"),
))
.expect("Send error");
// Wait permanently so message channel doesn't get disconnected
// Achieved differently in production
Ok(futures::Async::NotReady)
})
.map_err(|e: tokio::sync::mpsc::error::SendError| println!("Error: {:?}", e)),
);
Ok(())
}));
}
Repo
Using this version of your future shows the problem:
tokio::spawn(
future::poll_fn(move || {
eprintln!("Starting poll_fn");
let from_channel = rx
.poll()
.map_err(|_e| tokio::io::Error::new(tokio::io::ErrorKind::Other, "Poll error"));
if let Some((packet, to)) = futures::try_ready!(dbg!(from_channel)) {
futures::try_ready!(dbg!(sock.poll_send_to(packet.as_slice(), &to)));
}
Ok(futures::Async::Ready(()))
})
.map_err(|e: tokio::io::Error| println!("Error: {:?}", e)),
);
Here's the slightly cleaned-up output:
Starting poll_fn
[src/main.rs:21] from_channel = Ok(NotReady)
Starting poll_fn
[src/main.rs:21] from_channel = Ok(Ready(Some(/* ... */)))
[src/main.rs:22] sock.poll_send_to(packet.as_slice(), &to) = Ok(NotReady)
Starting poll_fn
[src/main.rs:21] from_channel = Ok(NotReady)
In words:
The future starts.
There's nothing ready from the channel; the channel registers a notification.
The future returns.
The channel gets a value and notifies the task.
The future starts again.
There's a value ready from the channel.
Sending on the socket is not ready; the socket registers a notification.
The future returns.
The socket is cleared and notifies the task.
The future starts again.
There's nothing ready from the channel; the channel registers a notification.
The future returns.
Nothing else is ever added to the channel.
In short, you aren't correctly maintaining your state machine inside of your future. You need to know how far you got the last time the future ran and start at that point the next time it ran.
There's a reason that the async / await syntax is much-anticipated: it will write these state machines for you.
I don't know why you've chosen to use the lower-level poll-based interface. I'd use the higher-level Future-based one:
tokio::spawn({
rx.fold(sock, |sock, (packet, to)| {
sock.send_dgram(packet, &to)
.inspect(|_| println!("Sent it!"))
.map(|(sock, _)| sock)
.map_err(|e| panic!("Error: {:?}", e))
})
.map(drop)
.map_err(|e| panic!("Error: {:?}", e))
});
the Future-based interface [...] destroys the socket(and buffer) on error
This is a good reason to use the poll-based interface, but I'd still just dip into it long enough to implement your own future. Something like this:
struct X(UdpSocket);
struct XSendGram<D> {
sock: Option<UdpSocket>,
data: D,
addr: SocketAddr,
}
impl X {
fn send_dgram<D>(self, data: D, addr: SocketAddr) -> XSendGram<D> {
XSendGram {
sock: Some(self.0),
data,
addr,
}
}
}
impl<D> Future for XSendGram<D>
where
D: AsRef<[u8]>,
{
type Item = (X, usize);
type Error = (X, std::io::Error);
fn poll(&mut self) -> Result<Async<Self::Item>, Self::Error> {
let mut sock = self.sock.take().expect("Future called after success or failure");
match sock.poll_send_to(self.data.as_ref(), &self.addr) {
Ok(Async::Ready(bytes)) => Ok(Async::Ready((X(sock), bytes))),
Ok(Async::NotReady) => {
self.sock = Some(sock); // Restore it for the next call
Ok(Async::NotReady)
}
Err(e) => Err((X(sock), e)),
}
}
}
tokio::spawn({
rx.fold(X(sock), |sock, (packet, to)| {
sock.send_dgram(packet, to)
.inspect(|(_, n)| println!("Sent {} bytes", n))
.then(|r| match r {
Ok((sock, _)) | Err((sock, _)) => future::ok(sock),
})
})
.map(drop)
.map_err(|e| panic!("Error: {:?}", e))
});

Resources