fix(protocol,edge): optimize tunnel frame handling and zero-copy uploads in edge I/O
This commit is contained in:
@@ -13,6 +13,15 @@ use serde::{Deserialize, Serialize};
|
||||
|
||||
use remoteingress_protocol::*;
|
||||
|
||||
type EdgeTlsStream = tokio_rustls::client::TlsStream<TcpStream>;
|
||||
|
||||
/// Result of processing a frame (shared with hub.rs pattern).
|
||||
#[allow(dead_code)]
|
||||
enum EdgeFrameAction {
|
||||
Continue,
|
||||
Disconnect(String),
|
||||
}
|
||||
|
||||
/// Per-stream state tracked in the edge's client_writers map.
|
||||
struct EdgeStreamState {
|
||||
/// Channel to deliver FRAME_DATA_BACK payloads to the hub_to_client task.
|
||||
@@ -272,6 +281,83 @@ enum EdgeLoopResult {
|
||||
Reconnect(String), // reason for disconnection
|
||||
}
|
||||
|
||||
/// Process a single frame received from the hub side of the tunnel.
|
||||
/// Handles FRAME_DATA_BACK, FRAME_WINDOW_UPDATE_BACK, FRAME_CLOSE_BACK, FRAME_CONFIG, FRAME_PING.
|
||||
async fn handle_edge_frame(
|
||||
frame: Frame,
|
||||
tunnel_io: &mut remoteingress_protocol::TunnelIo<EdgeTlsStream>,
|
||||
client_writers: &Arc<Mutex<HashMap<u32, EdgeStreamState>>>,
|
||||
listen_ports: &Arc<RwLock<Vec<u16>>>,
|
||||
event_tx: &mpsc::Sender<EdgeEvent>,
|
||||
tunnel_writer_tx: &mpsc::Sender<Vec<u8>>,
|
||||
tunnel_data_tx: &mpsc::Sender<Vec<u8>>,
|
||||
port_listeners: &mut HashMap<u16, JoinHandle<()>>,
|
||||
active_streams: &Arc<AtomicU32>,
|
||||
next_stream_id: &Arc<AtomicU32>,
|
||||
edge_id: &str,
|
||||
connection_token: &CancellationToken,
|
||||
bind_address: &str,
|
||||
) -> EdgeFrameAction {
|
||||
match frame.frame_type {
|
||||
FRAME_DATA_BACK => {
|
||||
let mut writers = client_writers.lock().await;
|
||||
if let Some(state) = writers.get(&frame.stream_id) {
|
||||
if state.back_tx.try_send(frame.payload).is_err() {
|
||||
log::warn!("Stream {} back-channel full, closing", frame.stream_id);
|
||||
writers.remove(&frame.stream_id);
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_WINDOW_UPDATE_BACK => {
|
||||
if let Some(increment) = decode_window_update(&frame.payload) {
|
||||
if increment > 0 {
|
||||
let writers = client_writers.lock().await;
|
||||
if let Some(state) = writers.get(&frame.stream_id) {
|
||||
let prev = state.send_window.fetch_add(increment, Ordering::Release);
|
||||
if prev + increment > MAX_WINDOW_SIZE {
|
||||
state.send_window.store(MAX_WINDOW_SIZE, Ordering::Release);
|
||||
}
|
||||
state.window_notify.notify_one();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_CLOSE_BACK => {
|
||||
let mut writers = client_writers.lock().await;
|
||||
writers.remove(&frame.stream_id);
|
||||
}
|
||||
FRAME_CONFIG => {
|
||||
if let Ok(update) = serde_json::from_slice::<ConfigUpdate>(&frame.payload) {
|
||||
log::info!("Config update from hub: ports {:?}", update.listen_ports);
|
||||
*listen_ports.write().await = update.listen_ports.clone();
|
||||
let _ = event_tx.try_send(EdgeEvent::PortsUpdated {
|
||||
listen_ports: update.listen_ports.clone(),
|
||||
});
|
||||
apply_port_config(
|
||||
&update.listen_ports,
|
||||
port_listeners,
|
||||
tunnel_writer_tx,
|
||||
tunnel_data_tx,
|
||||
client_writers,
|
||||
active_streams,
|
||||
next_stream_id,
|
||||
edge_id,
|
||||
connection_token,
|
||||
bind_address,
|
||||
);
|
||||
}
|
||||
}
|
||||
FRAME_PING => {
|
||||
// Queue PONG directly — no channel round-trip, guaranteed delivery
|
||||
tunnel_io.queue_ctrl(encode_frame(0, FRAME_PONG, &[]));
|
||||
}
|
||||
_ => {
|
||||
log::warn!("Unexpected frame type {} from hub", frame.frame_type);
|
||||
}
|
||||
}
|
||||
EdgeFrameAction::Continue
|
||||
}
|
||||
|
||||
async fn connect_to_hub_and_run(
|
||||
config: &EdgeConfig,
|
||||
connected: &Arc<RwLock<bool>>,
|
||||
@@ -436,73 +522,22 @@ async fn connect_to_hub_and_run(
|
||||
let result = 'io_loop: loop {
|
||||
// Drain any buffered frames
|
||||
loop {
|
||||
match tunnel_io.try_parse_frame() {
|
||||
Some(Ok(frame)) => {
|
||||
last_activity = Instant::now();
|
||||
liveness_deadline.as_mut().reset(last_activity + liveness_timeout_dur);
|
||||
match frame.frame_type {
|
||||
FRAME_DATA_BACK => {
|
||||
let mut writers = client_writers.lock().await;
|
||||
if let Some(state) = writers.get(&frame.stream_id) {
|
||||
if state.back_tx.try_send(frame.payload).is_err() {
|
||||
log::warn!("Stream {} back-channel full, closing", frame.stream_id);
|
||||
writers.remove(&frame.stream_id);
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_WINDOW_UPDATE_BACK => {
|
||||
if let Some(increment) = decode_window_update(&frame.payload) {
|
||||
if increment > 0 {
|
||||
let writers = client_writers.lock().await;
|
||||
if let Some(state) = writers.get(&frame.stream_id) {
|
||||
let prev = state.send_window.fetch_add(increment, Ordering::Release);
|
||||
if prev + increment > MAX_WINDOW_SIZE {
|
||||
state.send_window.store(MAX_WINDOW_SIZE, Ordering::Release);
|
||||
}
|
||||
state.window_notify.notify_one();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_CLOSE_BACK => {
|
||||
let mut writers = client_writers.lock().await;
|
||||
writers.remove(&frame.stream_id);
|
||||
}
|
||||
FRAME_CONFIG => {
|
||||
if let Ok(update) = serde_json::from_slice::<ConfigUpdate>(&frame.payload) {
|
||||
log::info!("Config update from hub: ports {:?}", update.listen_ports);
|
||||
*listen_ports.write().await = update.listen_ports.clone();
|
||||
let _ = event_tx.try_send(EdgeEvent::PortsUpdated {
|
||||
listen_ports: update.listen_ports.clone(),
|
||||
});
|
||||
apply_port_config(
|
||||
&update.listen_ports,
|
||||
&mut port_listeners,
|
||||
&tunnel_writer_tx,
|
||||
&tunnel_data_tx,
|
||||
&client_writers,
|
||||
active_streams,
|
||||
next_stream_id,
|
||||
&config.edge_id,
|
||||
connection_token,
|
||||
bind_address,
|
||||
);
|
||||
}
|
||||
}
|
||||
FRAME_PING => {
|
||||
// Queue PONG directly — no channel round-trip, guaranteed delivery
|
||||
tunnel_io.queue_ctrl(encode_frame(0, FRAME_PONG, &[]));
|
||||
}
|
||||
_ => {
|
||||
log::warn!("Unexpected frame type {} from hub", frame.frame_type);
|
||||
}
|
||||
}
|
||||
}
|
||||
let frame = match tunnel_io.try_parse_frame() {
|
||||
Some(Ok(f)) => f,
|
||||
Some(Err(e)) => {
|
||||
log::error!("Hub frame error: {}", e);
|
||||
break 'io_loop EdgeLoopResult::Reconnect(format!("hub_frame_error: {}", e));
|
||||
}
|
||||
None => break,
|
||||
};
|
||||
last_activity = Instant::now();
|
||||
liveness_deadline.as_mut().reset(last_activity + liveness_timeout_dur);
|
||||
if let EdgeFrameAction::Disconnect(reason) = handle_edge_frame(
|
||||
frame, &mut tunnel_io, &client_writers, listen_ports, event_tx,
|
||||
&tunnel_writer_tx, &tunnel_data_tx, &mut port_listeners,
|
||||
active_streams, next_stream_id, &config.edge_id, connection_token, bind_address,
|
||||
).await {
|
||||
break 'io_loop EdgeLoopResult::Reconnect(reason);
|
||||
}
|
||||
}
|
||||
|
||||
@@ -515,61 +550,12 @@ async fn connect_to_hub_and_run(
|
||||
remoteingress_protocol::TunnelEvent::Frame(frame) => {
|
||||
last_activity = Instant::now();
|
||||
liveness_deadline.as_mut().reset(last_activity + liveness_timeout_dur);
|
||||
match frame.frame_type {
|
||||
FRAME_DATA_BACK => {
|
||||
let mut writers = client_writers.lock().await;
|
||||
if let Some(state) = writers.get(&frame.stream_id) {
|
||||
if state.back_tx.try_send(frame.payload).is_err() {
|
||||
log::warn!("Stream {} back-channel full, closing", frame.stream_id);
|
||||
writers.remove(&frame.stream_id);
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_WINDOW_UPDATE_BACK => {
|
||||
if let Some(increment) = decode_window_update(&frame.payload) {
|
||||
if increment > 0 {
|
||||
let writers = client_writers.lock().await;
|
||||
if let Some(state) = writers.get(&frame.stream_id) {
|
||||
let prev = state.send_window.fetch_add(increment, Ordering::Release);
|
||||
if prev + increment > MAX_WINDOW_SIZE {
|
||||
state.send_window.store(MAX_WINDOW_SIZE, Ordering::Release);
|
||||
}
|
||||
state.window_notify.notify_one();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_CLOSE_BACK => {
|
||||
let mut writers = client_writers.lock().await;
|
||||
writers.remove(&frame.stream_id);
|
||||
}
|
||||
FRAME_CONFIG => {
|
||||
if let Ok(update) = serde_json::from_slice::<ConfigUpdate>(&frame.payload) {
|
||||
log::info!("Config update from hub: ports {:?}", update.listen_ports);
|
||||
*listen_ports.write().await = update.listen_ports.clone();
|
||||
let _ = event_tx.try_send(EdgeEvent::PortsUpdated {
|
||||
listen_ports: update.listen_ports.clone(),
|
||||
});
|
||||
apply_port_config(
|
||||
&update.listen_ports,
|
||||
&mut port_listeners,
|
||||
&tunnel_writer_tx,
|
||||
&tunnel_data_tx,
|
||||
&client_writers,
|
||||
active_streams,
|
||||
next_stream_id,
|
||||
&config.edge_id,
|
||||
connection_token,
|
||||
bind_address,
|
||||
);
|
||||
}
|
||||
}
|
||||
FRAME_PING => {
|
||||
tunnel_io.queue_ctrl(encode_frame(0, FRAME_PONG, &[]));
|
||||
}
|
||||
_ => {
|
||||
log::warn!("Unexpected frame type {} from hub", frame.frame_type);
|
||||
}
|
||||
if let EdgeFrameAction::Disconnect(reason) = handle_edge_frame(
|
||||
frame, &mut tunnel_io, &client_writers, listen_ports, event_tx,
|
||||
&tunnel_writer_tx, &tunnel_data_tx, &mut port_listeners,
|
||||
active_streams, next_stream_id, &config.edge_id, connection_token, bind_address,
|
||||
).await {
|
||||
break EdgeLoopResult::Reconnect(reason);
|
||||
}
|
||||
}
|
||||
remoteingress_protocol::TunnelEvent::Eof => {
|
||||
@@ -813,15 +799,21 @@ async fn handle_client_connection(
|
||||
let _ = client_write.shutdown().await;
|
||||
});
|
||||
|
||||
// Task: client -> hub (upload direction) with per-stream flow control
|
||||
let mut buf = vec![0u8; 32768];
|
||||
// Task: client -> hub (upload direction) with per-stream flow control.
|
||||
// Zero-copy: read payload directly after the header, then prepend header.
|
||||
let mut buf = vec![0u8; FRAME_HEADER_SIZE + 32768];
|
||||
loop {
|
||||
// Wait for send window to have capacity (with stall timeout)
|
||||
// Wait for send window to have capacity (with stall timeout).
|
||||
// Safe pattern: register notified BEFORE checking the condition
|
||||
// to avoid missing a notify_one that fires between load and select.
|
||||
loop {
|
||||
let notified = window_notify.notified();
|
||||
tokio::pin!(notified);
|
||||
notified.as_mut().enable();
|
||||
let w = send_window.load(Ordering::Acquire);
|
||||
if w > 0 { break; }
|
||||
tokio::select! {
|
||||
_ = window_notify.notified() => continue,
|
||||
_ = notified => continue,
|
||||
_ = client_token.cancelled() => break,
|
||||
_ = tokio::time::sleep(Duration::from_secs(120)) => {
|
||||
log::warn!("Stream {} upload stalled (window empty for 120s)", stream_id);
|
||||
@@ -844,15 +836,16 @@ async fn handle_client_connection(
|
||||
let adaptive_cap = remoteingress_protocol::compute_window_for_stream_count(
|
||||
active_streams.load(Ordering::Relaxed),
|
||||
) as usize;
|
||||
let max_read = w.min(buf.len()).min(adaptive_cap);
|
||||
let max_read = w.min(32768).min(adaptive_cap);
|
||||
|
||||
tokio::select! {
|
||||
read_result = client_read.read(&mut buf[..max_read]) => {
|
||||
read_result = client_read.read(&mut buf[FRAME_HEADER_SIZE..FRAME_HEADER_SIZE + max_read]) => {
|
||||
match read_result {
|
||||
Ok(0) => break,
|
||||
Ok(n) => {
|
||||
send_window.fetch_sub(n as u32, Ordering::Release);
|
||||
let data_frame = encode_frame(stream_id, FRAME_DATA, &buf[..n]);
|
||||
encode_frame_header(&mut buf, stream_id, FRAME_DATA, n);
|
||||
let data_frame = buf[..FRAME_HEADER_SIZE + n].to_vec();
|
||||
if tunnel_data_tx.send(data_frame).await.is_err() {
|
||||
log::warn!("Stream {} data channel closed, closing", stream_id);
|
||||
break;
|
||||
|
||||
@@ -12,6 +12,19 @@ use serde::{Deserialize, Serialize};
|
||||
|
||||
use remoteingress_protocol::*;
|
||||
|
||||
type HubTlsStream = tokio_rustls::server::TlsStream<TcpStream>;
|
||||
|
||||
/// Per-stream data channel capacity. With 4MB window and 32KB frames,
|
||||
/// at most ~128 frames are in-flight. 256 provides comfortable headroom.
|
||||
const PER_STREAM_DATA_CAPACITY: usize = 256;
|
||||
|
||||
/// Result of processing a frame.
|
||||
#[allow(dead_code)]
|
||||
enum FrameAction {
|
||||
Continue,
|
||||
Disconnect(String),
|
||||
}
|
||||
|
||||
/// Per-stream state tracked in the hub's stream map.
|
||||
struct HubStreamState {
|
||||
/// Channel to deliver FRAME_DATA payloads to the upstream writer task.
|
||||
@@ -123,7 +136,7 @@ pub struct TunnelHub {
|
||||
struct ConnectedEdgeInfo {
|
||||
connected_at: u64,
|
||||
peer_addr: String,
|
||||
active_streams: Arc<Mutex<HashMap<u32, HubStreamState>>>,
|
||||
edge_stream_count: Arc<AtomicU32>,
|
||||
config_tx: mpsc::Sender<EdgeConfigUpdate>,
|
||||
#[allow(dead_code)] // kept alive for Drop — cancels child tokens when edge is removed
|
||||
cancel_token: CancellationToken,
|
||||
@@ -189,11 +202,10 @@ impl TunnelHub {
|
||||
|
||||
let mut connected = Vec::new();
|
||||
for (id, info) in edges.iter() {
|
||||
let streams = info.active_streams.lock().await;
|
||||
connected.push(ConnectedEdgeStatus {
|
||||
edge_id: id.clone(),
|
||||
connected_at: info.connected_at,
|
||||
active_streams: streams.len(),
|
||||
active_streams: info.edge_stream_count.load(Ordering::Relaxed) as usize,
|
||||
peer_addr: info.peer_addr.clone(),
|
||||
});
|
||||
}
|
||||
@@ -287,6 +299,285 @@ impl Drop for TunnelHub {
|
||||
/// Maximum concurrent streams per edge connection.
|
||||
const MAX_STREAMS_PER_EDGE: usize = 1024;
|
||||
|
||||
/// Process a single frame received from the edge side of the tunnel.
|
||||
/// Handles FRAME_OPEN, FRAME_DATA, FRAME_WINDOW_UPDATE, FRAME_CLOSE, and FRAME_PONG.
|
||||
async fn handle_hub_frame(
|
||||
frame: Frame,
|
||||
tunnel_io: &mut remoteingress_protocol::TunnelIo<HubTlsStream>,
|
||||
streams: &mut HashMap<u32, HubStreamState>,
|
||||
stream_semaphore: &Arc<Semaphore>,
|
||||
edge_stream_count: &Arc<AtomicU32>,
|
||||
edge_id: &str,
|
||||
event_tx: &mpsc::Sender<HubEvent>,
|
||||
ctrl_tx: &mpsc::Sender<Vec<u8>>,
|
||||
data_tx: &mpsc::Sender<Vec<u8>>,
|
||||
target_host: &str,
|
||||
edge_token: &CancellationToken,
|
||||
cleanup_tx: &mpsc::Sender<u32>,
|
||||
) -> FrameAction {
|
||||
match frame.frame_type {
|
||||
FRAME_OPEN => {
|
||||
// A4: Check stream limit before processing
|
||||
let permit = match stream_semaphore.clone().try_acquire_owned() {
|
||||
Ok(p) => p,
|
||||
Err(_) => {
|
||||
log::warn!("Edge {} exceeded max streams ({}), rejecting stream {}",
|
||||
edge_id, MAX_STREAMS_PER_EDGE, frame.stream_id);
|
||||
let close_frame = encode_frame(frame.stream_id, FRAME_CLOSE_BACK, &[]);
|
||||
tunnel_io.queue_ctrl(close_frame);
|
||||
return FrameAction::Continue;
|
||||
}
|
||||
};
|
||||
|
||||
// Payload is PROXY v1 header line
|
||||
let proxy_header = String::from_utf8_lossy(&frame.payload).to_string();
|
||||
|
||||
// Parse destination port from PROXY header
|
||||
let dest_port = parse_dest_port_from_proxy(&proxy_header).unwrap_or(443);
|
||||
|
||||
let stream_id = frame.stream_id;
|
||||
let cleanup = cleanup_tx.clone();
|
||||
let writer_tx = ctrl_tx.clone(); // control: CLOSE_BACK, WINDOW_UPDATE_BACK
|
||||
let data_writer_tx = data_tx.clone(); // data: DATA_BACK
|
||||
let target = target_host.to_string();
|
||||
let stream_token = edge_token.child_token();
|
||||
|
||||
let _ = event_tx.try_send(HubEvent::StreamOpened {
|
||||
edge_id: edge_id.to_string(),
|
||||
stream_id,
|
||||
});
|
||||
|
||||
// Create channel for data from edge to this stream
|
||||
let (stream_data_tx, mut stream_data_rx) = mpsc::channel::<Vec<u8>>(PER_STREAM_DATA_CAPACITY);
|
||||
// Adaptive initial window: scale with current stream count
|
||||
// to keep total in-flight data within the 32MB budget.
|
||||
let initial_window = compute_window_for_stream_count(
|
||||
edge_stream_count.load(Ordering::Relaxed),
|
||||
);
|
||||
let send_window = Arc::new(AtomicU32::new(initial_window));
|
||||
let window_notify = Arc::new(Notify::new());
|
||||
streams.insert(stream_id, HubStreamState {
|
||||
data_tx: stream_data_tx,
|
||||
cancel_token: stream_token.clone(),
|
||||
send_window: Arc::clone(&send_window),
|
||||
window_notify: Arc::clone(&window_notify),
|
||||
});
|
||||
|
||||
// Spawn task: connect to SmartProxy, send PROXY header, pipe data
|
||||
let stream_counter = Arc::clone(edge_stream_count);
|
||||
tokio::spawn(async move {
|
||||
let _permit = permit; // hold semaphore permit until stream completes
|
||||
stream_counter.fetch_add(1, Ordering::Relaxed);
|
||||
|
||||
let result = async {
|
||||
// A2: Connect to SmartProxy with timeout
|
||||
let mut upstream = tokio::time::timeout(
|
||||
Duration::from_secs(10),
|
||||
TcpStream::connect((target.as_str(), dest_port)),
|
||||
)
|
||||
.await
|
||||
.map_err(|_| -> Box<dyn std::error::Error + Send + Sync> {
|
||||
format!("connect to SmartProxy {}:{} timed out (10s)", target, dest_port).into()
|
||||
})??;
|
||||
|
||||
upstream.set_nodelay(true)?;
|
||||
upstream.write_all(proxy_header.as_bytes()).await?;
|
||||
|
||||
let (mut up_read, mut up_write) =
|
||||
upstream.into_split();
|
||||
|
||||
// Forward data from edge (via channel) to SmartProxy
|
||||
// After writing to upstream, send WINDOW_UPDATE_BACK to edge
|
||||
let writer_token = stream_token.clone();
|
||||
let wub_tx = writer_tx.clone();
|
||||
let stream_counter_w = Arc::clone(&stream_counter);
|
||||
let writer_for_edge_data = tokio::spawn(async move {
|
||||
let mut consumed_since_update: u32 = 0;
|
||||
loop {
|
||||
tokio::select! {
|
||||
data = stream_data_rx.recv() => {
|
||||
match data {
|
||||
Some(data) => {
|
||||
let len = data.len() as u32;
|
||||
// Check cancellation alongside the write so we respond
|
||||
// promptly to FRAME_CLOSE instead of blocking up to 60s.
|
||||
let write_result = tokio::select! {
|
||||
r = tokio::time::timeout(
|
||||
Duration::from_secs(60),
|
||||
up_write.write_all(&data),
|
||||
) => r,
|
||||
_ = writer_token.cancelled() => break,
|
||||
};
|
||||
match write_result {
|
||||
Ok(Ok(())) => {}
|
||||
Ok(Err(_)) => break,
|
||||
Err(_) => {
|
||||
log::warn!("Stream {} write to upstream timed out (60s)", stream_id);
|
||||
break;
|
||||
}
|
||||
}
|
||||
// Track consumption for adaptive flow control.
|
||||
// Increment capped to adaptive window to limit per-stream in-flight data.
|
||||
consumed_since_update += len;
|
||||
let adaptive_window = remoteingress_protocol::compute_window_for_stream_count(
|
||||
stream_counter_w.load(Ordering::Relaxed),
|
||||
);
|
||||
let threshold = adaptive_window / 2;
|
||||
if consumed_since_update >= threshold {
|
||||
let increment = consumed_since_update.min(adaptive_window);
|
||||
let frame = encode_window_update(stream_id, FRAME_WINDOW_UPDATE_BACK, increment);
|
||||
if wub_tx.try_send(frame).is_ok() {
|
||||
consumed_since_update -= increment;
|
||||
}
|
||||
// If try_send fails, keep accumulating — retry on next threshold
|
||||
}
|
||||
}
|
||||
None => break,
|
||||
}
|
||||
}
|
||||
_ = writer_token.cancelled() => break,
|
||||
}
|
||||
}
|
||||
// Send final window update for remaining consumed bytes
|
||||
if consumed_since_update > 0 {
|
||||
let frame = encode_window_update(stream_id, FRAME_WINDOW_UPDATE_BACK, consumed_since_update);
|
||||
let _ = wub_tx.try_send(frame);
|
||||
}
|
||||
let _ = up_write.shutdown().await;
|
||||
});
|
||||
|
||||
// Forward data from SmartProxy back to edge via writer channel
|
||||
// with per-stream flow control (check send_window before reading).
|
||||
// Zero-copy: read payload directly after the header, then prepend header.
|
||||
let mut buf = vec![0u8; FRAME_HEADER_SIZE + 32768];
|
||||
loop {
|
||||
// Wait for send window to have capacity (with stall timeout).
|
||||
// Safe pattern: register notified BEFORE checking the condition
|
||||
// to avoid missing a notify_one that fires between load and select.
|
||||
loop {
|
||||
let notified = window_notify.notified();
|
||||
tokio::pin!(notified);
|
||||
notified.as_mut().enable();
|
||||
let w = send_window.load(Ordering::Acquire);
|
||||
if w > 0 { break; }
|
||||
tokio::select! {
|
||||
_ = notified => continue,
|
||||
_ = stream_token.cancelled() => break,
|
||||
_ = tokio::time::sleep(Duration::from_secs(120)) => {
|
||||
log::warn!("Stream {} download stalled (window empty for 120s)", stream_id);
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
if stream_token.is_cancelled() { break; }
|
||||
|
||||
// Limit read size to available window.
|
||||
// IMPORTANT: if window is 0 (stall timeout fired), we must NOT
|
||||
// read into an empty buffer — read(&mut buf[..0]) returns Ok(0)
|
||||
// which would be falsely interpreted as EOF.
|
||||
let w = send_window.load(Ordering::Acquire) as usize;
|
||||
if w == 0 {
|
||||
log::warn!("Stream {} download: window still 0 after stall timeout, closing", stream_id);
|
||||
break;
|
||||
}
|
||||
// Adaptive: cap read to current per-stream target window
|
||||
let adaptive_cap = remoteingress_protocol::compute_window_for_stream_count(
|
||||
stream_counter.load(Ordering::Relaxed),
|
||||
) as usize;
|
||||
let max_read = w.min(32768).min(adaptive_cap);
|
||||
|
||||
tokio::select! {
|
||||
read_result = up_read.read(&mut buf[FRAME_HEADER_SIZE..FRAME_HEADER_SIZE + max_read]) => {
|
||||
match read_result {
|
||||
Ok(0) => break,
|
||||
Ok(n) => {
|
||||
send_window.fetch_sub(n as u32, Ordering::Release);
|
||||
encode_frame_header(&mut buf, stream_id, FRAME_DATA_BACK, n);
|
||||
let frame = buf[..FRAME_HEADER_SIZE + n].to_vec();
|
||||
if data_writer_tx.send(frame).await.is_err() {
|
||||
log::warn!("Stream {} data channel closed, closing", stream_id);
|
||||
break;
|
||||
}
|
||||
}
|
||||
Err(_) => break,
|
||||
}
|
||||
}
|
||||
_ = stream_token.cancelled() => break,
|
||||
}
|
||||
}
|
||||
|
||||
// Send CLOSE_BACK via DATA channel (must arrive AFTER last DATA_BACK).
|
||||
// Use send().await to guarantee delivery (try_send silently drops if full).
|
||||
if !stream_token.is_cancelled() {
|
||||
let close_frame = encode_frame(stream_id, FRAME_CLOSE_BACK, &[]);
|
||||
let _ = data_writer_tx.send(close_frame).await;
|
||||
}
|
||||
|
||||
writer_for_edge_data.abort();
|
||||
Ok::<(), Box<dyn std::error::Error + Send + Sync>>(())
|
||||
}
|
||||
.await;
|
||||
|
||||
if let Err(e) = result {
|
||||
log::error!("Stream {} error: {}", stream_id, e);
|
||||
// Send CLOSE_BACK via DATA channel on error (must arrive after any DATA_BACK).
|
||||
// Use send().await to guarantee delivery.
|
||||
if !stream_token.is_cancelled() {
|
||||
let close_frame = encode_frame(stream_id, FRAME_CLOSE_BACK, &[]);
|
||||
let _ = data_writer_tx.send(close_frame).await;
|
||||
}
|
||||
}
|
||||
|
||||
// Signal main loop to remove stream from the map
|
||||
let _ = cleanup.send(stream_id).await;
|
||||
stream_counter.fetch_sub(1, Ordering::Relaxed);
|
||||
});
|
||||
}
|
||||
FRAME_DATA => {
|
||||
// Non-blocking dispatch to per-stream channel.
|
||||
// With flow control, the sender should rarely exceed the channel capacity.
|
||||
if let Some(state) = streams.get(&frame.stream_id) {
|
||||
if state.data_tx.try_send(frame.payload).is_err() {
|
||||
log::warn!("Stream {} data channel full, closing stream", frame.stream_id);
|
||||
if let Some(state) = streams.remove(&frame.stream_id) {
|
||||
state.cancel_token.cancel();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_WINDOW_UPDATE => {
|
||||
// Edge consumed data — increase our send window for this stream
|
||||
if let Some(increment) = decode_window_update(&frame.payload) {
|
||||
if increment > 0 {
|
||||
if let Some(state) = streams.get(&frame.stream_id) {
|
||||
let prev = state.send_window.fetch_add(increment, Ordering::Release);
|
||||
if prev + increment > MAX_WINDOW_SIZE {
|
||||
state.send_window.store(MAX_WINDOW_SIZE, Ordering::Release);
|
||||
}
|
||||
state.window_notify.notify_one();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_CLOSE => {
|
||||
if let Some(state) = streams.remove(&frame.stream_id) {
|
||||
state.cancel_token.cancel();
|
||||
let _ = event_tx.try_send(HubEvent::StreamClosed {
|
||||
edge_id: edge_id.to_string(),
|
||||
stream_id: frame.stream_id,
|
||||
});
|
||||
}
|
||||
}
|
||||
FRAME_PONG => {
|
||||
log::debug!("Received PONG from edge {}", edge_id);
|
||||
}
|
||||
_ => {
|
||||
log::warn!("Unexpected frame type {} from edge", frame.frame_type);
|
||||
}
|
||||
}
|
||||
FrameAction::Continue
|
||||
}
|
||||
|
||||
/// Handle a single edge connection: authenticate, then enter frame loop.
|
||||
async fn handle_edge_connection(
|
||||
stream: TcpStream,
|
||||
@@ -368,8 +659,11 @@ async fn handle_edge_connection(
|
||||
tls_stream.flush().await?;
|
||||
|
||||
// Track this edge
|
||||
let streams: Arc<Mutex<HashMap<u32, HubStreamState>>> =
|
||||
Arc::new(Mutex::new(HashMap::new()));
|
||||
let mut streams: HashMap<u32, HubStreamState> = HashMap::new();
|
||||
// Per-edge active stream counter for adaptive flow control
|
||||
let edge_stream_count = Arc::new(AtomicU32::new(0));
|
||||
// Cleanup channel: spawned stream tasks send stream_id here when done
|
||||
let (cleanup_tx, mut cleanup_rx) = mpsc::channel::<u32>(256);
|
||||
let now = std::time::SystemTime::now()
|
||||
.duration_since(std::time::UNIX_EPOCH)
|
||||
.unwrap_or_default()
|
||||
@@ -385,16 +679,13 @@ async fn handle_edge_connection(
|
||||
ConnectedEdgeInfo {
|
||||
connected_at: now,
|
||||
peer_addr,
|
||||
active_streams: streams.clone(),
|
||||
edge_stream_count: edge_stream_count.clone(),
|
||||
config_tx,
|
||||
cancel_token: edge_token.clone(),
|
||||
},
|
||||
);
|
||||
}
|
||||
|
||||
// Per-edge active stream counter for adaptive flow control
|
||||
let edge_stream_count = Arc::new(AtomicU32::new(0));
|
||||
|
||||
// QoS dual-channel: ctrl frames have priority over data frames.
|
||||
// Stream handlers send through these channels -> TunnelIo drains them.
|
||||
let (ctrl_tx, mut ctrl_rx) = mpsc::channel::<Vec<u8>>(256);
|
||||
@@ -444,292 +735,36 @@ async fn handle_edge_connection(
|
||||
let mut disconnect_reason = "unknown".to_string();
|
||||
|
||||
'hub_loop: loop {
|
||||
// Drain completed stream cleanups from spawned tasks
|
||||
while let Ok(stream_id) = cleanup_rx.try_recv() {
|
||||
if streams.remove(&stream_id).is_some() {
|
||||
let _ = event_tx.try_send(HubEvent::StreamClosed {
|
||||
edge_id: edge_id.clone(),
|
||||
stream_id,
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
// Drain any buffered frames
|
||||
loop {
|
||||
match tunnel_io.try_parse_frame() {
|
||||
Some(Ok(frame)) => {
|
||||
// Reset liveness on any received frame
|
||||
last_activity = Instant::now();
|
||||
liveness_deadline.as_mut().reset(last_activity + liveness_timeout_dur);
|
||||
|
||||
match frame.frame_type {
|
||||
FRAME_OPEN => {
|
||||
// A4: Check stream limit before processing
|
||||
let permit = match stream_semaphore.clone().try_acquire_owned() {
|
||||
Ok(p) => p,
|
||||
Err(_) => {
|
||||
log::warn!("Edge {} exceeded max streams ({}), rejecting stream {}",
|
||||
edge_id, MAX_STREAMS_PER_EDGE, frame.stream_id);
|
||||
let close_frame = encode_frame(frame.stream_id, FRAME_CLOSE_BACK, &[]);
|
||||
tunnel_io.queue_ctrl(close_frame);
|
||||
continue;
|
||||
}
|
||||
};
|
||||
|
||||
// Payload is PROXY v1 header line
|
||||
let proxy_header = String::from_utf8_lossy(&frame.payload).to_string();
|
||||
|
||||
// Parse destination port from PROXY header
|
||||
let dest_port = parse_dest_port_from_proxy(&proxy_header).unwrap_or(443);
|
||||
|
||||
let stream_id = frame.stream_id;
|
||||
let edge_id_clone = edge_id.clone();
|
||||
let event_tx_clone = event_tx.clone();
|
||||
let streams_clone = streams.clone();
|
||||
let writer_tx = ctrl_tx.clone(); // control: CLOSE_BACK, WINDOW_UPDATE_BACK
|
||||
let data_writer_tx = data_tx.clone(); // data: DATA_BACK
|
||||
let target = target_host.clone();
|
||||
let stream_token = edge_token.child_token();
|
||||
|
||||
let _ = event_tx.try_send(HubEvent::StreamOpened {
|
||||
edge_id: edge_id.clone(),
|
||||
stream_id,
|
||||
});
|
||||
|
||||
// Create channel for data from edge to this stream (capacity 16 is sufficient with flow control)
|
||||
let (data_tx, mut data_rx) = mpsc::channel::<Vec<u8>>(1024);
|
||||
// Adaptive initial window: scale with current stream count
|
||||
// to keep total in-flight data within the 32MB budget.
|
||||
let initial_window = compute_window_for_stream_count(
|
||||
edge_stream_count.load(Ordering::Relaxed),
|
||||
);
|
||||
let send_window = Arc::new(AtomicU32::new(initial_window));
|
||||
let window_notify = Arc::new(Notify::new());
|
||||
{
|
||||
let mut s = streams.lock().await;
|
||||
s.insert(stream_id, HubStreamState {
|
||||
data_tx,
|
||||
cancel_token: stream_token.clone(),
|
||||
send_window: Arc::clone(&send_window),
|
||||
window_notify: Arc::clone(&window_notify),
|
||||
});
|
||||
}
|
||||
|
||||
// Spawn task: connect to SmartProxy, send PROXY header, pipe data
|
||||
let stream_counter = Arc::clone(&edge_stream_count);
|
||||
tokio::spawn(async move {
|
||||
let _permit = permit; // hold semaphore permit until stream completes
|
||||
stream_counter.fetch_add(1, Ordering::Relaxed);
|
||||
|
||||
let result = async {
|
||||
// A2: Connect to SmartProxy with timeout
|
||||
let mut upstream = tokio::time::timeout(
|
||||
Duration::from_secs(10),
|
||||
TcpStream::connect((target.as_str(), dest_port)),
|
||||
)
|
||||
.await
|
||||
.map_err(|_| -> Box<dyn std::error::Error + Send + Sync> {
|
||||
format!("connect to SmartProxy {}:{} timed out (10s)", target, dest_port).into()
|
||||
})??;
|
||||
|
||||
upstream.set_nodelay(true)?;
|
||||
upstream.write_all(proxy_header.as_bytes()).await?;
|
||||
|
||||
let (mut up_read, mut up_write) =
|
||||
upstream.into_split();
|
||||
|
||||
// Forward data from edge (via channel) to SmartProxy
|
||||
// After writing to upstream, send WINDOW_UPDATE_BACK to edge
|
||||
let writer_token = stream_token.clone();
|
||||
let wub_tx = writer_tx.clone();
|
||||
let stream_counter_w = Arc::clone(&stream_counter);
|
||||
let writer_for_edge_data = tokio::spawn(async move {
|
||||
let mut consumed_since_update: u32 = 0;
|
||||
loop {
|
||||
tokio::select! {
|
||||
data = data_rx.recv() => {
|
||||
match data {
|
||||
Some(data) => {
|
||||
let len = data.len() as u32;
|
||||
// Check cancellation alongside the write so we respond
|
||||
// promptly to FRAME_CLOSE instead of blocking up to 60s.
|
||||
let write_result = tokio::select! {
|
||||
r = tokio::time::timeout(
|
||||
Duration::from_secs(60),
|
||||
up_write.write_all(&data),
|
||||
) => r,
|
||||
_ = writer_token.cancelled() => break,
|
||||
};
|
||||
match write_result {
|
||||
Ok(Ok(())) => {}
|
||||
Ok(Err(_)) => break,
|
||||
Err(_) => {
|
||||
log::warn!("Stream {} write to upstream timed out (60s)", stream_id);
|
||||
break;
|
||||
}
|
||||
}
|
||||
// Track consumption for adaptive flow control.
|
||||
// Increment capped to adaptive window to limit per-stream in-flight data.
|
||||
consumed_since_update += len;
|
||||
let adaptive_window = remoteingress_protocol::compute_window_for_stream_count(
|
||||
stream_counter_w.load(Ordering::Relaxed),
|
||||
);
|
||||
let threshold = adaptive_window / 2;
|
||||
if consumed_since_update >= threshold {
|
||||
let increment = consumed_since_update.min(adaptive_window);
|
||||
let frame = encode_window_update(stream_id, FRAME_WINDOW_UPDATE_BACK, increment);
|
||||
if wub_tx.try_send(frame).is_ok() {
|
||||
consumed_since_update -= increment;
|
||||
}
|
||||
// If try_send fails, keep accumulating — retry on next threshold
|
||||
}
|
||||
}
|
||||
None => break,
|
||||
}
|
||||
}
|
||||
_ = writer_token.cancelled() => break,
|
||||
}
|
||||
}
|
||||
// Send final window update for remaining consumed bytes
|
||||
if consumed_since_update > 0 {
|
||||
let frame = encode_window_update(stream_id, FRAME_WINDOW_UPDATE_BACK, consumed_since_update);
|
||||
let _ = wub_tx.try_send(frame);
|
||||
}
|
||||
let _ = up_write.shutdown().await;
|
||||
});
|
||||
|
||||
// Forward data from SmartProxy back to edge via writer channel
|
||||
// with per-stream flow control (check send_window before reading)
|
||||
let mut buf = vec![0u8; 32768];
|
||||
loop {
|
||||
// Wait for send window to have capacity (with stall timeout)
|
||||
loop {
|
||||
let w = send_window.load(Ordering::Acquire);
|
||||
if w > 0 { break; }
|
||||
tokio::select! {
|
||||
_ = window_notify.notified() => continue,
|
||||
_ = stream_token.cancelled() => break,
|
||||
_ = tokio::time::sleep(Duration::from_secs(120)) => {
|
||||
log::warn!("Stream {} download stalled (window empty for 120s)", stream_id);
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
if stream_token.is_cancelled() { break; }
|
||||
|
||||
// Limit read size to available window.
|
||||
// IMPORTANT: if window is 0 (stall timeout fired), we must NOT
|
||||
// read into an empty buffer — read(&mut buf[..0]) returns Ok(0)
|
||||
// which would be falsely interpreted as EOF.
|
||||
let w = send_window.load(Ordering::Acquire) as usize;
|
||||
if w == 0 {
|
||||
log::warn!("Stream {} download: window still 0 after stall timeout, closing", stream_id);
|
||||
break;
|
||||
}
|
||||
// Adaptive: cap read to current per-stream target window
|
||||
let adaptive_cap = remoteingress_protocol::compute_window_for_stream_count(
|
||||
stream_counter.load(Ordering::Relaxed),
|
||||
) as usize;
|
||||
let max_read = w.min(buf.len()).min(adaptive_cap);
|
||||
|
||||
tokio::select! {
|
||||
read_result = up_read.read(&mut buf[..max_read]) => {
|
||||
match read_result {
|
||||
Ok(0) => break,
|
||||
Ok(n) => {
|
||||
send_window.fetch_sub(n as u32, Ordering::Release);
|
||||
let frame =
|
||||
encode_frame(stream_id, FRAME_DATA_BACK, &buf[..n]);
|
||||
if data_writer_tx.send(frame).await.is_err() {
|
||||
log::warn!("Stream {} data channel closed, closing", stream_id);
|
||||
break;
|
||||
}
|
||||
}
|
||||
Err(_) => break,
|
||||
}
|
||||
}
|
||||
_ = stream_token.cancelled() => break,
|
||||
}
|
||||
}
|
||||
|
||||
// Send CLOSE_BACK via DATA channel (must arrive AFTER last DATA_BACK).
|
||||
// Use send().await to guarantee delivery (try_send silently drops if full).
|
||||
if !stream_token.is_cancelled() {
|
||||
let close_frame = encode_frame(stream_id, FRAME_CLOSE_BACK, &[]);
|
||||
let _ = data_writer_tx.send(close_frame).await;
|
||||
}
|
||||
|
||||
writer_for_edge_data.abort();
|
||||
Ok::<(), Box<dyn std::error::Error + Send + Sync>>(())
|
||||
}
|
||||
.await;
|
||||
|
||||
if let Err(e) = result {
|
||||
log::error!("Stream {} error: {}", stream_id, e);
|
||||
// Send CLOSE_BACK via DATA channel on error (must arrive after any DATA_BACK).
|
||||
// Use send().await to guarantee delivery.
|
||||
if !stream_token.is_cancelled() {
|
||||
let close_frame = encode_frame(stream_id, FRAME_CLOSE_BACK, &[]);
|
||||
let _ = data_writer_tx.send(close_frame).await;
|
||||
}
|
||||
}
|
||||
|
||||
// Clean up stream (guard against duplicate if FRAME_CLOSE already removed it)
|
||||
let was_present = {
|
||||
let mut s = streams_clone.lock().await;
|
||||
s.remove(&stream_id).is_some()
|
||||
};
|
||||
if was_present {
|
||||
let _ = event_tx_clone.try_send(HubEvent::StreamClosed {
|
||||
edge_id: edge_id_clone,
|
||||
stream_id,
|
||||
});
|
||||
}
|
||||
stream_counter.fetch_sub(1, Ordering::Relaxed);
|
||||
});
|
||||
}
|
||||
FRAME_DATA => {
|
||||
// Non-blocking dispatch to per-stream channel.
|
||||
// With flow control, the sender should rarely exceed the channel capacity.
|
||||
let mut s = streams.lock().await;
|
||||
if let Some(state) = s.get(&frame.stream_id) {
|
||||
if state.data_tx.try_send(frame.payload).is_err() {
|
||||
log::warn!("Stream {} data channel full, closing stream", frame.stream_id);
|
||||
if let Some(state) = s.remove(&frame.stream_id) {
|
||||
state.cancel_token.cancel();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_WINDOW_UPDATE => {
|
||||
// Edge consumed data — increase our send window for this stream
|
||||
if let Some(increment) = decode_window_update(&frame.payload) {
|
||||
if increment > 0 {
|
||||
let s = streams.lock().await;
|
||||
if let Some(state) = s.get(&frame.stream_id) {
|
||||
let prev = state.send_window.fetch_add(increment, Ordering::Release);
|
||||
if prev + increment > MAX_WINDOW_SIZE {
|
||||
state.send_window.store(MAX_WINDOW_SIZE, Ordering::Release);
|
||||
}
|
||||
state.window_notify.notify_one();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_CLOSE => {
|
||||
let mut s = streams.lock().await;
|
||||
if let Some(state) = s.remove(&frame.stream_id) {
|
||||
state.cancel_token.cancel();
|
||||
let _ = event_tx.try_send(HubEvent::StreamClosed {
|
||||
edge_id: edge_id.clone(),
|
||||
stream_id: frame.stream_id,
|
||||
});
|
||||
}
|
||||
}
|
||||
FRAME_PONG => {
|
||||
log::debug!("Received PONG from edge {}", edge_id);
|
||||
}
|
||||
_ => {
|
||||
log::warn!("Unexpected frame type {} from edge", frame.frame_type);
|
||||
}
|
||||
}
|
||||
}
|
||||
let frame = match tunnel_io.try_parse_frame() {
|
||||
Some(Ok(f)) => f,
|
||||
Some(Err(e)) => {
|
||||
log::error!("Edge {} frame error: {}", edge_id, e);
|
||||
disconnect_reason = format!("edge_frame_error: {}", e);
|
||||
break 'hub_loop;
|
||||
}
|
||||
None => break,
|
||||
};
|
||||
last_activity = Instant::now();
|
||||
liveness_deadline.as_mut().reset(last_activity + liveness_timeout_dur);
|
||||
if let FrameAction::Disconnect(reason) = handle_hub_frame(
|
||||
frame, &mut tunnel_io, &mut streams, &stream_semaphore, &edge_stream_count,
|
||||
&edge_id, &event_tx, &ctrl_tx, &data_tx, &target_host, &edge_token,
|
||||
&cleanup_tx,
|
||||
).await {
|
||||
disconnect_reason = reason;
|
||||
break 'hub_loop;
|
||||
}
|
||||
}
|
||||
|
||||
@@ -744,280 +779,15 @@ async fn handle_edge_connection(
|
||||
|
||||
match event {
|
||||
remoteingress_protocol::TunnelEvent::Frame(frame) => {
|
||||
// Reset liveness on any received frame
|
||||
last_activity = Instant::now();
|
||||
liveness_deadline.as_mut().reset(last_activity + liveness_timeout_dur);
|
||||
|
||||
match frame.frame_type {
|
||||
FRAME_OPEN => {
|
||||
// A4: Check stream limit before processing
|
||||
let permit = match stream_semaphore.clone().try_acquire_owned() {
|
||||
Ok(p) => p,
|
||||
Err(_) => {
|
||||
log::warn!("Edge {} exceeded max streams ({}), rejecting stream {}",
|
||||
edge_id, MAX_STREAMS_PER_EDGE, frame.stream_id);
|
||||
let close_frame = encode_frame(frame.stream_id, FRAME_CLOSE_BACK, &[]);
|
||||
tunnel_io.queue_ctrl(close_frame);
|
||||
continue;
|
||||
}
|
||||
};
|
||||
|
||||
// Payload is PROXY v1 header line
|
||||
let proxy_header = String::from_utf8_lossy(&frame.payload).to_string();
|
||||
|
||||
// Parse destination port from PROXY header
|
||||
let dest_port = parse_dest_port_from_proxy(&proxy_header).unwrap_or(443);
|
||||
|
||||
let stream_id = frame.stream_id;
|
||||
let edge_id_clone = edge_id.clone();
|
||||
let event_tx_clone = event_tx.clone();
|
||||
let streams_clone = streams.clone();
|
||||
let writer_tx = ctrl_tx.clone(); // control: CLOSE_BACK, WINDOW_UPDATE_BACK
|
||||
let data_writer_tx = data_tx.clone(); // data: DATA_BACK
|
||||
let target = target_host.clone();
|
||||
let stream_token = edge_token.child_token();
|
||||
|
||||
let _ = event_tx.try_send(HubEvent::StreamOpened {
|
||||
edge_id: edge_id.clone(),
|
||||
stream_id,
|
||||
});
|
||||
|
||||
// Create channel for data from edge to this stream (capacity 16 is sufficient with flow control)
|
||||
let (data_tx, mut data_rx) = mpsc::channel::<Vec<u8>>(256);
|
||||
// Adaptive initial window: scale with current stream count
|
||||
// to keep total in-flight data within the 32MB budget.
|
||||
let initial_window = compute_window_for_stream_count(
|
||||
edge_stream_count.load(Ordering::Relaxed),
|
||||
);
|
||||
let send_window = Arc::new(AtomicU32::new(initial_window));
|
||||
let window_notify = Arc::new(Notify::new());
|
||||
{
|
||||
let mut s = streams.lock().await;
|
||||
s.insert(stream_id, HubStreamState {
|
||||
data_tx,
|
||||
cancel_token: stream_token.clone(),
|
||||
send_window: Arc::clone(&send_window),
|
||||
window_notify: Arc::clone(&window_notify),
|
||||
});
|
||||
}
|
||||
|
||||
// Spawn task: connect to SmartProxy, send PROXY header, pipe data
|
||||
let stream_counter = Arc::clone(&edge_stream_count);
|
||||
tokio::spawn(async move {
|
||||
let _permit = permit; // hold semaphore permit until stream completes
|
||||
stream_counter.fetch_add(1, Ordering::Relaxed);
|
||||
|
||||
let result = async {
|
||||
// A2: Connect to SmartProxy with timeout
|
||||
let mut upstream = tokio::time::timeout(
|
||||
Duration::from_secs(10),
|
||||
TcpStream::connect((target.as_str(), dest_port)),
|
||||
)
|
||||
.await
|
||||
.map_err(|_| -> Box<dyn std::error::Error + Send + Sync> {
|
||||
format!("connect to SmartProxy {}:{} timed out (10s)", target, dest_port).into()
|
||||
})??;
|
||||
|
||||
upstream.set_nodelay(true)?;
|
||||
upstream.write_all(proxy_header.as_bytes()).await?;
|
||||
|
||||
let (mut up_read, mut up_write) =
|
||||
upstream.into_split();
|
||||
|
||||
// Forward data from edge (via channel) to SmartProxy
|
||||
// After writing to upstream, send WINDOW_UPDATE_BACK to edge
|
||||
let writer_token = stream_token.clone();
|
||||
let wub_tx = writer_tx.clone();
|
||||
let stream_counter_w = Arc::clone(&stream_counter);
|
||||
let writer_for_edge_data = tokio::spawn(async move {
|
||||
let mut consumed_since_update: u32 = 0;
|
||||
loop {
|
||||
tokio::select! {
|
||||
data = data_rx.recv() => {
|
||||
match data {
|
||||
Some(data) => {
|
||||
let len = data.len() as u32;
|
||||
// Check cancellation alongside the write so we respond
|
||||
// promptly to FRAME_CLOSE instead of blocking up to 60s.
|
||||
let write_result = tokio::select! {
|
||||
r = tokio::time::timeout(
|
||||
Duration::from_secs(60),
|
||||
up_write.write_all(&data),
|
||||
) => r,
|
||||
_ = writer_token.cancelled() => break,
|
||||
};
|
||||
match write_result {
|
||||
Ok(Ok(())) => {}
|
||||
Ok(Err(_)) => break,
|
||||
Err(_) => {
|
||||
log::warn!("Stream {} write to upstream timed out (60s)", stream_id);
|
||||
break;
|
||||
}
|
||||
}
|
||||
// Track consumption for adaptive flow control.
|
||||
// Increment capped to adaptive window to limit per-stream in-flight data.
|
||||
consumed_since_update += len;
|
||||
let adaptive_window = remoteingress_protocol::compute_window_for_stream_count(
|
||||
stream_counter_w.load(Ordering::Relaxed),
|
||||
);
|
||||
let threshold = adaptive_window / 2;
|
||||
if consumed_since_update >= threshold {
|
||||
let increment = consumed_since_update.min(adaptive_window);
|
||||
let frame = encode_window_update(stream_id, FRAME_WINDOW_UPDATE_BACK, increment);
|
||||
if wub_tx.try_send(frame).is_ok() {
|
||||
consumed_since_update -= increment;
|
||||
}
|
||||
// If try_send fails, keep accumulating — retry on next threshold
|
||||
}
|
||||
}
|
||||
None => break,
|
||||
}
|
||||
}
|
||||
_ = writer_token.cancelled() => break,
|
||||
}
|
||||
}
|
||||
// Send final window update for remaining consumed bytes
|
||||
if consumed_since_update > 0 {
|
||||
let frame = encode_window_update(stream_id, FRAME_WINDOW_UPDATE_BACK, consumed_since_update);
|
||||
let _ = wub_tx.try_send(frame);
|
||||
}
|
||||
let _ = up_write.shutdown().await;
|
||||
});
|
||||
|
||||
// Forward data from SmartProxy back to edge via writer channel
|
||||
// with per-stream flow control (check send_window before reading)
|
||||
let mut buf = vec![0u8; 32768];
|
||||
loop {
|
||||
// Wait for send window to have capacity (with stall timeout)
|
||||
loop {
|
||||
let w = send_window.load(Ordering::Acquire);
|
||||
if w > 0 { break; }
|
||||
tokio::select! {
|
||||
_ = window_notify.notified() => continue,
|
||||
_ = stream_token.cancelled() => break,
|
||||
_ = tokio::time::sleep(Duration::from_secs(120)) => {
|
||||
log::warn!("Stream {} download stalled (window empty for 120s)", stream_id);
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
if stream_token.is_cancelled() { break; }
|
||||
|
||||
// Limit read size to available window.
|
||||
// IMPORTANT: if window is 0 (stall timeout fired), we must NOT
|
||||
// read into an empty buffer — read(&mut buf[..0]) returns Ok(0)
|
||||
// which would be falsely interpreted as EOF.
|
||||
let w = send_window.load(Ordering::Acquire) as usize;
|
||||
if w == 0 {
|
||||
log::warn!("Stream {} download: window still 0 after stall timeout, closing", stream_id);
|
||||
break;
|
||||
}
|
||||
// Adaptive: cap read to current per-stream target window
|
||||
let adaptive_cap = remoteingress_protocol::compute_window_for_stream_count(
|
||||
stream_counter.load(Ordering::Relaxed),
|
||||
) as usize;
|
||||
let max_read = w.min(buf.len()).min(adaptive_cap);
|
||||
|
||||
tokio::select! {
|
||||
read_result = up_read.read(&mut buf[..max_read]) => {
|
||||
match read_result {
|
||||
Ok(0) => break,
|
||||
Ok(n) => {
|
||||
send_window.fetch_sub(n as u32, Ordering::Release);
|
||||
let frame =
|
||||
encode_frame(stream_id, FRAME_DATA_BACK, &buf[..n]);
|
||||
if data_writer_tx.send(frame).await.is_err() {
|
||||
log::warn!("Stream {} data channel closed, closing", stream_id);
|
||||
break;
|
||||
}
|
||||
}
|
||||
Err(_) => break,
|
||||
}
|
||||
}
|
||||
_ = stream_token.cancelled() => break,
|
||||
}
|
||||
}
|
||||
|
||||
// Send CLOSE_BACK via DATA channel (must arrive AFTER last DATA_BACK).
|
||||
// Use send().await to guarantee delivery (try_send silently drops if full).
|
||||
if !stream_token.is_cancelled() {
|
||||
let close_frame = encode_frame(stream_id, FRAME_CLOSE_BACK, &[]);
|
||||
let _ = data_writer_tx.send(close_frame).await;
|
||||
}
|
||||
|
||||
writer_for_edge_data.abort();
|
||||
Ok::<(), Box<dyn std::error::Error + Send + Sync>>(())
|
||||
}
|
||||
.await;
|
||||
|
||||
if let Err(e) = result {
|
||||
log::error!("Stream {} error: {}", stream_id, e);
|
||||
// Send CLOSE_BACK via DATA channel on error (must arrive after any DATA_BACK).
|
||||
// Use send().await to guarantee delivery.
|
||||
if !stream_token.is_cancelled() {
|
||||
let close_frame = encode_frame(stream_id, FRAME_CLOSE_BACK, &[]);
|
||||
let _ = data_writer_tx.send(close_frame).await;
|
||||
}
|
||||
}
|
||||
|
||||
// Clean up stream (guard against duplicate if FRAME_CLOSE already removed it)
|
||||
let was_present = {
|
||||
let mut s = streams_clone.lock().await;
|
||||
s.remove(&stream_id).is_some()
|
||||
};
|
||||
if was_present {
|
||||
let _ = event_tx_clone.try_send(HubEvent::StreamClosed {
|
||||
edge_id: edge_id_clone,
|
||||
stream_id,
|
||||
});
|
||||
}
|
||||
stream_counter.fetch_sub(1, Ordering::Relaxed);
|
||||
});
|
||||
}
|
||||
FRAME_DATA => {
|
||||
// Non-blocking dispatch to per-stream channel.
|
||||
// With flow control, the sender should rarely exceed the channel capacity.
|
||||
let mut s = streams.lock().await;
|
||||
if let Some(state) = s.get(&frame.stream_id) {
|
||||
if state.data_tx.try_send(frame.payload).is_err() {
|
||||
log::warn!("Stream {} data channel full, closing stream", frame.stream_id);
|
||||
if let Some(state) = s.remove(&frame.stream_id) {
|
||||
state.cancel_token.cancel();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_WINDOW_UPDATE => {
|
||||
// Edge consumed data — increase our send window for this stream
|
||||
if let Some(increment) = decode_window_update(&frame.payload) {
|
||||
if increment > 0 {
|
||||
let s = streams.lock().await;
|
||||
if let Some(state) = s.get(&frame.stream_id) {
|
||||
let prev = state.send_window.fetch_add(increment, Ordering::Release);
|
||||
if prev + increment > MAX_WINDOW_SIZE {
|
||||
state.send_window.store(MAX_WINDOW_SIZE, Ordering::Release);
|
||||
}
|
||||
state.window_notify.notify_one();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
FRAME_CLOSE => {
|
||||
let mut s = streams.lock().await;
|
||||
if let Some(state) = s.remove(&frame.stream_id) {
|
||||
state.cancel_token.cancel();
|
||||
let _ = event_tx.try_send(HubEvent::StreamClosed {
|
||||
edge_id: edge_id.clone(),
|
||||
stream_id: frame.stream_id,
|
||||
});
|
||||
}
|
||||
}
|
||||
FRAME_PONG => {
|
||||
log::debug!("Received PONG from edge {}", edge_id);
|
||||
}
|
||||
_ => {
|
||||
log::warn!("Unexpected frame type {} from edge", frame.frame_type);
|
||||
}
|
||||
if let FrameAction::Disconnect(reason) = handle_hub_frame(
|
||||
frame, &mut tunnel_io, &mut streams, &stream_semaphore, &edge_stream_count,
|
||||
&edge_id, &event_tx, &ctrl_tx, &data_tx, &target_host, &edge_token,
|
||||
&cleanup_tx,
|
||||
).await {
|
||||
disconnect_reason = reason;
|
||||
break;
|
||||
}
|
||||
}
|
||||
remoteingress_protocol::TunnelEvent::Eof => {
|
||||
|
||||
Reference in New Issue
Block a user