Compare commits

...

6 Commits

Author SHA1 Message Date
Alexey 3bd5637e47 ME + Admission + Cleanup Correctness: merge pull request #779 from telemt/flow
ME + Admission + Cleanup Correctness
2026-05-10 14:23:09 +03:00
Alexey 57b2aa0453 Rustfmt 2026-05-10 14:14:52 +03:00
Alexey 10c7cb2e0c Middle Relay Cancellation Errors 2026-05-10 14:12:15 +03:00
Alexey 900b574fb8 Harden ME Writer Cancellation paths 2026-05-10 14:09:10 +03:00
Alexey beed6b4679 Middle Wait Deadlines + Tighten Session Release State 2026-05-10 13:58:02 +03:00
Alexey eef2a38c75 Type Route Cutovers + Reduce IP Tracker cleanup pressure 2026-05-10 13:55:01 +03:00
21 changed files with 523 additions and 188 deletions
Generated
+3 -3
View File
@@ -2404,9 +2404,9 @@ checksum = "f87165f0995f63a9fbeea62b64d10b4d9d8e78ec6d7d51fb2125fda7bb36788f"
[[package]] [[package]]
name = "rustls-webpki" name = "rustls-webpki"
version = "0.103.12" version = "0.103.13"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8279bb85272c9f10811ae6a6c547ff594d6a7f3c6c6b02ee9726d1d0dcfcdd06" checksum = "61c429a8649f110dddef65e2a5ad240f747e85f7758a6bccc7e5777bd33f756e"
dependencies = [ dependencies = [
"aws-lc-rs", "aws-lc-rs",
"ring", "ring",
@@ -2791,7 +2791,7 @@ checksum = "7b2093cf4c8eb1e67749a6762251bc9cd836b6fc171623bd0a9d324d37af2417"
[[package]] [[package]]
name = "telemt" name = "telemt"
version = "3.4.10" version = "3.4.11"
dependencies = [ dependencies = [
"aes", "aes",
"anyhow", "anyhow",
+1 -1
View File
@@ -1,6 +1,6 @@
[package] [package]
name = "telemt" name = "telemt"
version = "3.4.10" version = "3.4.11"
edition = "2024" edition = "2024"
[features] [features]
+1 -4
View File
@@ -621,10 +621,7 @@ fn collect_unknown_config_keys(parsed_toml: &toml::Value) -> Vec<UnknownConfigKe
} }
} }
if let Some(upstreams) = parsed_toml if let Some(upstreams) = parsed_toml.get("upstreams").and_then(toml::Value::as_array) {
.get("upstreams")
.and_then(toml::Value::as_array)
{
for (idx, upstream) in upstreams.iter().enumerate() { for (idx, upstream) in upstreams.iter().enumerate() {
check_nested_table_value( check_nested_table_value(
&mut unknown, &mut unknown,
+12
View File
@@ -225,6 +225,18 @@ pub enum ProxyError {
#[error("ME connection lost")] #[error("ME connection lost")]
MiddleConnectionLost, MiddleConnectionLost,
#[error("Session terminated")]
RouteSwitched,
#[error("Traffic budget wait cancelled")]
TrafficBudgetWaitCancelled,
#[error("Traffic budget wait deadline exceeded")]
TrafficBudgetWaitDeadlineExceeded,
#[error("ME client writer cancelled")]
MiddleClientWriterCancelled,
// ============= Config Errors ============= // ============= Config Errors =============
#[error("Config error: {0}")] #[error("Config error: {0}")]
Config(String), Config(String),
+23 -5
View File
@@ -32,6 +32,7 @@ pub struct UserIpTracker {
limit_mode: Arc<RwLock<UserMaxUniqueIpsMode>>, limit_mode: Arc<RwLock<UserMaxUniqueIpsMode>>,
limit_window: Arc<RwLock<Duration>>, limit_window: Arc<RwLock<Duration>>,
last_compact_epoch_secs: Arc<AtomicU64>, last_compact_epoch_secs: Arc<AtomicU64>,
cleanup_queue_len: Arc<AtomicU64>,
cleanup_queue: Arc<Mutex<HashMap<(String, IpAddr), usize>>>, cleanup_queue: Arc<Mutex<HashMap<(String, IpAddr), usize>>>,
cleanup_drain_lock: Arc<AsyncMutex<()>>, cleanup_drain_lock: Arc<AsyncMutex<()>>,
} }
@@ -72,6 +73,7 @@ impl UserIpTracker {
limit_mode: Arc::new(RwLock::new(UserMaxUniqueIpsMode::ActiveWindow)), limit_mode: Arc::new(RwLock::new(UserMaxUniqueIpsMode::ActiveWindow)),
limit_window: Arc::new(RwLock::new(Duration::from_secs(30))), limit_window: Arc::new(RwLock::new(Duration::from_secs(30))),
last_compact_epoch_secs: Arc::new(AtomicU64::new(0)), last_compact_epoch_secs: Arc::new(AtomicU64::new(0)),
cleanup_queue_len: Arc::new(AtomicU64::new(0)),
cleanup_queue: Arc::new(Mutex::new(HashMap::new())), cleanup_queue: Arc::new(Mutex::new(HashMap::new())),
cleanup_drain_lock: Arc::new(AsyncMutex::new(())), cleanup_drain_lock: Arc::new(AsyncMutex::new(())),
} }
@@ -120,6 +122,9 @@ impl UserIpTracker {
match self.cleanup_queue.lock() { match self.cleanup_queue.lock() {
Ok(mut queue) => { Ok(mut queue) => {
let count = queue.entry((user, ip)).or_insert(0); let count = queue.entry((user, ip)).or_insert(0);
if *count == 0 {
self.cleanup_queue_len.fetch_add(1, Ordering::Relaxed);
}
*count = count.saturating_add(1); *count = count.saturating_add(1);
self.cleanup_deferred_releases self.cleanup_deferred_releases
.fetch_add(1, Ordering::Relaxed); .fetch_add(1, Ordering::Relaxed);
@@ -127,6 +132,9 @@ impl UserIpTracker {
Err(poisoned) => { Err(poisoned) => {
let mut queue = poisoned.into_inner(); let mut queue = poisoned.into_inner();
let count = queue.entry((user.clone(), ip)).or_insert(0); let count = queue.entry((user.clone(), ip)).or_insert(0);
if *count == 0 {
self.cleanup_queue_len.fetch_add(1, Ordering::Relaxed);
}
*count = count.saturating_add(1); *count = count.saturating_add(1);
self.cleanup_deferred_releases self.cleanup_deferred_releases
.fetch_add(1, Ordering::Relaxed); .fetch_add(1, Ordering::Relaxed);
@@ -156,6 +164,9 @@ impl UserIpTracker {
} }
pub(crate) async fn drain_cleanup_queue(&self) { pub(crate) async fn drain_cleanup_queue(&self) {
if self.cleanup_queue_len.load(Ordering::Relaxed) == 0 {
return;
}
let Ok(_drain_guard) = self.cleanup_drain_lock.try_lock() else { let Ok(_drain_guard) = self.cleanup_drain_lock.try_lock() else {
return; return;
}; };
@@ -173,6 +184,7 @@ impl UserIpTracker {
break; break;
}; };
if let Some(count) = queue.remove(&key) { if let Some(count) = queue.remove(&key) {
self.cleanup_queue_len.fetch_sub(1, Ordering::Relaxed);
drained.insert(key, count); drained.insert(key, count);
} }
} }
@@ -191,6 +203,7 @@ impl UserIpTracker {
break; break;
}; };
if let Some(count) = queue.remove(&key) { if let Some(count) = queue.remove(&key) {
self.cleanup_queue_len.fetch_sub(1, Ordering::Relaxed);
drained.insert(key, count); drained.insert(key, count);
} }
} }
@@ -294,12 +307,17 @@ impl UserIpTracker {
} }
} }
pub async fn run_periodic_maintenance(self: Arc<Self>) {
let mut interval = tokio::time::interval(Duration::from_secs(1));
loop {
interval.tick().await;
self.drain_cleanup_queue().await;
self.maybe_compact_empty_users().await;
}
}
pub async fn memory_stats(&self) -> UserIpTrackerMemoryStats { pub async fn memory_stats(&self) -> UserIpTrackerMemoryStats {
let cleanup_queue_len = self let cleanup_queue_len = self.cleanup_queue_len.load(Ordering::Relaxed) as usize;
.cleanup_queue
.lock()
.unwrap_or_else(|poisoned| poisoned.into_inner())
.len();
let active_ips = self.active_ips.read().await; let active_ips = self.active_ips.read().await;
let recent_ips = self.recent_ips.read().await; let recent_ips = self.recent_ips.read().await;
let active_entries = active_ips.values().map(HashMap::len).sum(); let active_entries = active_ips.values().map(HashMap::len).sum();
+5 -9
View File
@@ -13,7 +13,7 @@ use crate::config::{ProxyConfig, RstOnCloseMode};
use crate::crypto::SecureRandom; use crate::crypto::SecureRandom;
use crate::ip_tracker::UserIpTracker; use crate::ip_tracker::UserIpTracker;
use crate::proxy::ClientHandler; use crate::proxy::ClientHandler;
use crate::proxy::route_mode::{ROUTE_SWITCH_ERROR_MSG, RouteRuntimeController}; use crate::proxy::route_mode::RouteRuntimeController;
use crate::proxy::shared_state::ProxySharedState; use crate::proxy::shared_state::ProxySharedState;
use crate::startup::{COMPONENT_LISTENERS_BIND, StartupTracker}; use crate::startup::{COMPONENT_LISTENERS_BIND, StartupTracker};
use crate::stats::beobachten::BeobachtenStore; use crate::stats::beobachten::BeobachtenStore;
@@ -492,14 +492,10 @@ pub(crate) fn spawn_tcp_accept_loops(
let handshake_close_reason = let handshake_close_reason =
expected_handshake_close_description(&e); expected_handshake_close_description(&e);
let me_closed = matches!( let me_closed =
&e, matches!(&e, crate::error::ProxyError::MiddleConnectionLost);
crate::error::ProxyError::MiddleConnectionLost let route_switched =
); matches!(&e, crate::error::ProxyError::RouteSwitched);
let route_switched = matches!(
&e,
crate::error::ProxyError::Proxy(msg) if msg == ROUTE_SWITCH_ERROR_MSG
);
match (peer_close_reason, me_closed) { match (peer_close_reason, me_closed) {
(Some(reason), _) => { (Some(reason), _) => {
+8 -1
View File
@@ -75,7 +75,14 @@ pub(crate) async fn spawn_runtime_tasks(
let stats_maintenance = stats.clone(); let stats_maintenance = stats.clone();
tokio::spawn(async move { tokio::spawn(async move {
stats_maintenance.run_periodic_user_stats_maintenance().await; stats_maintenance
.run_periodic_user_stats_maintenance()
.await;
});
let ip_tracker_maintenance = ip_tracker.clone();
tokio::spawn(async move {
ip_tracker_maintenance.run_periodic_maintenance().await;
}); });
let detected_ip_v4: Option<IpAddr> = probe.detected_ipv4.map(IpAddr::V4); let detected_ip_v4: Option<IpAddr> = probe.detected_ipv4.map(IpAddr::V4);
+9 -2
View File
@@ -8,8 +8,8 @@
//! //!
//! SIGHUP is handled separately in config/hot_reload.rs for config reload. //! SIGHUP is handled separately in config/hot_reload.rs for config reload.
use std::sync::Arc;
use std::path::PathBuf; use std::path::PathBuf;
use std::sync::Arc;
use std::time::{Duration, Instant}; use std::time::{Duration, Instant};
#[cfg(not(unix))] #[cfg(not(unix))]
@@ -52,7 +52,14 @@ pub(crate) async fn wait_for_shutdown(
quota_state_path: PathBuf, quota_state_path: PathBuf,
) { ) {
let signal = wait_for_shutdown_signal().await; let signal = wait_for_shutdown_signal().await;
perform_shutdown(signal, process_started_at, me_pool, &stats, quota_state_path).await; perform_shutdown(
signal,
process_started_at,
me_pool,
&stats,
quota_state_path,
)
.await;
} }
/// Waits for any shutdown signal (SIGINT, SIGTERM, SIGQUIT). /// Waits for any shutdown signal (SIGINT, SIGTERM, SIGQUIT).
+20 -33
View File
@@ -374,10 +374,7 @@ async fn render_tls_front_profile_health(
out, out,
"# HELP telemt_tls_front_profile_age_seconds Age of cached TLS front profile data per configured domain" "# HELP telemt_tls_front_profile_age_seconds Age of cached TLS front profile data per configured domain"
); );
let _ = writeln!( let _ = writeln!(out, "# TYPE telemt_tls_front_profile_age_seconds gauge");
out,
"# TYPE telemt_tls_front_profile_age_seconds gauge"
);
let _ = writeln!( let _ = writeln!(
out, out,
"# HELP telemt_tls_front_profile_app_data_records TLS front cached app-data record count per configured domain" "# HELP telemt_tls_front_profile_app_data_records TLS front cached app-data record count per configured domain"
@@ -390,10 +387,7 @@ async fn render_tls_front_profile_health(
out, out,
"# HELP telemt_tls_front_profile_ticket_records TLS front cached ticket-like tail record count per configured domain" "# HELP telemt_tls_front_profile_ticket_records TLS front cached ticket-like tail record count per configured domain"
); );
let _ = writeln!( let _ = writeln!(out, "# TYPE telemt_tls_front_profile_ticket_records gauge");
out,
"# TYPE telemt_tls_front_profile_ticket_records gauge"
);
let _ = writeln!( let _ = writeln!(
out, out,
"# HELP telemt_tls_front_profile_change_cipher_spec_records TLS front cached ChangeCipherSpec record count per configured domain" "# HELP telemt_tls_front_profile_change_cipher_spec_records TLS front cached ChangeCipherSpec record count per configured domain"
@@ -406,21 +400,14 @@ async fn render_tls_front_profile_health(
out, out,
"# HELP telemt_tls_front_profile_app_data_bytes TLS front cached total app-data bytes per configured domain" "# HELP telemt_tls_front_profile_app_data_bytes TLS front cached total app-data bytes per configured domain"
); );
let _ = writeln!( let _ = writeln!(out, "# TYPE telemt_tls_front_profile_app_data_bytes gauge");
out,
"# TYPE telemt_tls_front_profile_app_data_bytes gauge"
);
for item in health { for item in health {
let domain = prometheus_label_value(&item.domain); let domain = prometheus_label_value(&item.domain);
let _ = writeln!( let _ = writeln!(
out, out,
"telemt_tls_front_profile_info{{domain=\"{}\",source=\"{}\",is_default=\"{}\",has_cert_info=\"{}\",has_cert_payload=\"{}\"}} 1", "telemt_tls_front_profile_info{{domain=\"{}\",source=\"{}\",is_default=\"{}\",has_cert_info=\"{}\",has_cert_payload=\"{}\"}} 1",
domain, domain, item.source, item.is_default, item.has_cert_info, item.has_cert_payload
item.source,
item.is_default,
item.has_cert_info,
item.has_cert_payload
); );
let _ = writeln!( let _ = writeln!(
out, out,
@@ -771,10 +758,7 @@ async fn render_metrics(
out, out,
"# HELP telemt_quota_contention_timeout_total Quota reservations that hit the bounded contention budget" "# HELP telemt_quota_contention_timeout_total Quota reservations that hit the bounded contention budget"
); );
let _ = writeln!( let _ = writeln!(out, "# TYPE telemt_quota_contention_timeout_total counter");
out,
"# TYPE telemt_quota_contention_timeout_total counter"
);
let _ = writeln!( let _ = writeln!(
out, out,
"telemt_quota_contention_timeout_total {}", "telemt_quota_contention_timeout_total {}",
@@ -917,10 +901,7 @@ async fn render_metrics(
out, out,
"# HELP telemt_rate_limiter_burst_bound_bytes Configured upper bound for one direct relay rate-limit burst" "# HELP telemt_rate_limiter_burst_bound_bytes Configured upper bound for one direct relay rate-limit burst"
); );
let _ = writeln!( let _ = writeln!(out, "# TYPE telemt_rate_limiter_burst_bound_bytes gauge");
out,
"# TYPE telemt_rate_limiter_burst_bound_bytes gauge"
);
let _ = writeln!( let _ = writeln!(
out, out,
"telemt_rate_limiter_burst_bound_bytes{{direction=\"up\"}} {}", "telemt_rate_limiter_burst_bound_bytes{{direction=\"up\"}} {}",
@@ -3793,8 +3774,9 @@ mod tests {
))); )));
assert!(output.contains("telemt_connections_total 2")); assert!(output.contains("telemt_connections_total 2"));
assert!(output.contains("telemt_connections_bad_total 1")); assert!(output.contains("telemt_connections_bad_total 1"));
assert!(output assert!(output.contains(
.contains("telemt_connections_bad_by_class_total{class=\"tls_handshake_bad_client\"} 1")); "telemt_connections_bad_by_class_total{class=\"tls_handshake_bad_client\"} 1"
));
assert!(output.contains("telemt_handshake_timeouts_total 1")); assert!(output.contains("telemt_handshake_timeouts_total 1"));
assert!(output.contains("telemt_handshake_failures_by_class_total{class=\"timeout\"} 1")); assert!(output.contains("telemt_handshake_failures_by_class_total{class=\"timeout\"} 1"));
assert!(output.contains("telemt_auth_expensive_checks_total 9")); assert!(output.contains("telemt_auth_expensive_checks_total 9"));
@@ -3910,16 +3892,21 @@ mod tests {
output.contains("telemt_tls_front_profile_info{domain=\"fallback.example\",source=\"default\",is_default=\"true\",has_cert_info=\"false\",has_cert_payload=\"false\"} 1") output.contains("telemt_tls_front_profile_info{domain=\"fallback.example\",source=\"default\",is_default=\"true\",has_cert_info=\"false\",has_cert_payload=\"false\"} 1")
); );
assert!( assert!(
output.contains("telemt_tls_front_profile_app_data_records{domain=\"primary.example\"} 2") output.contains(
"telemt_tls_front_profile_app_data_records{domain=\"primary.example\"} 2"
)
); );
assert!( assert!(
output.contains("telemt_tls_front_profile_ticket_records{domain=\"primary.example\"} 1") output
.contains("telemt_tls_front_profile_ticket_records{domain=\"primary.example\"} 1")
); );
assert!(output.contains(
"telemt_tls_front_profile_change_cipher_spec_records{domain=\"primary.example\"} 1"
));
assert!( assert!(
output.contains("telemt_tls_front_profile_change_cipher_spec_records{domain=\"primary.example\"} 1") output.contains(
); "telemt_tls_front_profile_app_data_bytes{domain=\"primary.example\"} 1536"
assert!( )
output.contains("telemt_tls_front_profile_app_data_bytes{domain=\"primary.example\"} 1536")
); );
} }
+11 -10
View File
@@ -59,24 +59,30 @@ impl UserConnectionReservation {
} }
} }
async fn release(mut self) { fn mark_released(&mut self) -> bool {
if self.state != SessionReservationState::Active { if self.state != SessionReservationState::Active {
return false;
}
self.state = SessionReservationState::Released;
true
}
async fn release(mut self) {
if !self.mark_released() {
return; return;
} }
if self.tracks_ip { if self.tracks_ip {
self.ip_tracker.remove_ip(&self.user, self.ip).await; self.ip_tracker.remove_ip(&self.user, self.ip).await;
} }
self.state = SessionReservationState::Released;
self.stats.decrement_user_curr_connects(&self.user); self.stats.decrement_user_curr_connects(&self.user);
} }
} }
impl Drop for UserConnectionReservation { impl Drop for UserConnectionReservation {
fn drop(&mut self) { fn drop(&mut self) {
if self.state != SessionReservationState::Active { if !self.mark_released() {
return; return;
} }
self.state = SessionReservationState::Released;
self.stats.increment_session_drop_fallback_total(); self.stats.increment_session_drop_fallback_total();
self.stats.decrement_user_curr_connects(&self.user); self.stats.decrement_user_curr_connects(&self.user);
if self.tracks_ip { if self.tracks_ip {
@@ -996,12 +1002,7 @@ impl RunningClientHandler {
trusted = ?self.config.server.proxy_protocol_trusted_cidrs, trusted = ?self.config.server.proxy_protocol_trusted_cidrs,
"Rejecting PROXY protocol header from untrusted source" "Rejecting PROXY protocol header from untrusted source"
); );
record_beobachten_class( record_beobachten_class(&self.beobachten, &self.config, self.peer.ip(), "other");
&self.beobachten,
&self.config,
self.peer.ip(),
"other",
);
return Err(ProxyError::InvalidProxyProtocol); return Err(ProxyError::InvalidProxyProtocol);
} }
+2 -3
View File
@@ -18,8 +18,7 @@ use crate::error::{ProxyError, Result};
use crate::protocol::constants::*; use crate::protocol::constants::*;
use crate::proxy::handshake::{HandshakeSuccess, encrypt_tg_nonce_with_ciphers, generate_tg_nonce}; use crate::proxy::handshake::{HandshakeSuccess, encrypt_tg_nonce_with_ciphers, generate_tg_nonce};
use crate::proxy::route_mode::{ use crate::proxy::route_mode::{
ROUTE_SWITCH_ERROR_MSG, RelayRouteMode, RouteCutoverState, affected_cutover_state, RelayRouteMode, RouteCutoverState, affected_cutover_state, cutover_stagger_delay,
cutover_stagger_delay,
}; };
use crate::proxy::shared_state::{ use crate::proxy::shared_state::{
ConntrackCloseEvent, ConntrackClosePublishResult, ConntrackCloseReason, ProxySharedState, ConntrackCloseEvent, ConntrackClosePublishResult, ConntrackCloseReason, ProxySharedState,
@@ -360,7 +359,7 @@ where
"Cutover affected direct session, closing client connection" "Cutover affected direct session, closing client connection"
); );
tokio::time::sleep(delay).await; tokio::time::sleep(delay).await;
break Err(ProxyError::Proxy(ROUTE_SWITCH_ERROR_MSG.to_string())); break Err(ProxyError::RouteSwitched);
} }
tokio::select! { tokio::select! {
result = &mut relay_result => { result = &mut relay_result => {
+4 -1
View File
@@ -1901,7 +1901,10 @@ where
.auth_expensive_checks_total .auth_expensive_checks_total
.fetch_add(validation_checks as u64, Ordering::Relaxed); .fetch_add(validation_checks as u64, Ordering::Relaxed);
if config.access.is_user_source_ip_denied(user.as_str(), peer.ip()) { if config
.access
.is_user_source_ip_denied(user.as_str(), peer.ip())
{
auth_probe_record_failure_in(shared, peer.ip(), Instant::now()); auth_probe_record_failure_in(shared, peer.ip(), Instant::now());
maybe_apply_server_hello_delay(config).await; maybe_apply_server_hello_delay(config).await;
warn!( warn!(
+121 -74
View File
@@ -23,8 +23,7 @@ use crate::error::{ProxyError, Result};
use crate::protocol::constants::{secure_padding_len, *}; use crate::protocol::constants::{secure_padding_len, *};
use crate::proxy::handshake::HandshakeSuccess; use crate::proxy::handshake::HandshakeSuccess;
use crate::proxy::route_mode::{ use crate::proxy::route_mode::{
ROUTE_SWITCH_ERROR_MSG, RelayRouteMode, RouteCutoverState, affected_cutover_state, RelayRouteMode, RouteCutoverState, affected_cutover_state, cutover_stagger_delay,
cutover_stagger_delay,
}; };
use crate::proxy::shared_state::{ use crate::proxy::shared_state::{
ConntrackCloseEvent, ConntrackClosePublishResult, ConntrackCloseReason, ProxySharedState, ConntrackCloseEvent, ConntrackClosePublishResult, ConntrackCloseReason, ProxySharedState,
@@ -73,6 +72,7 @@ enum MiddleQuotaReserveError {
LimitExceeded, LimitExceeded,
Contended, Contended,
Cancelled, Cancelled,
DeadlineExceeded,
} }
#[derive(Default)] #[derive(Default)]
@@ -633,6 +633,7 @@ async fn reserve_user_quota_with_yield(
limit: u64, limit: u64,
stats: &Stats, stats: &Stats,
cancel: &CancellationToken, cancel: &CancellationToken,
deadline: Option<Instant>,
) -> std::result::Result<u64, MiddleQuotaReserveError> { ) -> std::result::Result<u64, MiddleQuotaReserveError> {
let mut backoff_ms = QUOTA_RESERVE_BACKOFF_MIN_MS; let mut backoff_ms = QUOTA_RESERVE_BACKOFF_MIN_MS;
let mut backoff_rounds = 0usize; let mut backoff_rounds = 0usize;
@@ -651,6 +652,10 @@ async fn reserve_user_quota_with_yield(
} }
tokio::task::yield_now().await; tokio::task::yield_now().await;
if deadline.is_some_and(|deadline| Instant::now() >= deadline) {
stats.increment_quota_contention_timeout_total();
return Err(MiddleQuotaReserveError::DeadlineExceeded);
}
tokio::select! { tokio::select! {
_ = tokio::time::sleep(Duration::from_millis(backoff_ms)) => {} _ = tokio::time::sleep(Duration::from_millis(backoff_ms)) => {}
_ = cancel.cancelled() => { _ = cancel.cancelled() => {
@@ -673,6 +678,7 @@ async fn wait_for_traffic_budget(
lease: Option<&Arc<TrafficLease>>, lease: Option<&Arc<TrafficLease>>,
direction: RateDirection, direction: RateDirection,
bytes: u64, bytes: u64,
deadline: Option<Instant>,
) -> Result<()> { ) -> Result<()> {
if bytes == 0 { if bytes == 0 {
return Ok(()); return Ok(());
@@ -690,6 +696,9 @@ async fn wait_for_traffic_budget(
} }
let wait_started_at = Instant::now(); let wait_started_at = Instant::now();
if deadline.is_some_and(|deadline| wait_started_at >= deadline) {
return Err(ProxyError::TrafficBudgetWaitDeadlineExceeded);
}
tokio::time::sleep(next_refill_delay()).await; tokio::time::sleep(next_refill_delay()).await;
let wait_ms = wait_started_at let wait_ms = wait_started_at
.elapsed() .elapsed()
@@ -712,6 +721,7 @@ async fn wait_for_traffic_budget_or_cancel(
bytes: u64, bytes: u64,
cancel: &CancellationToken, cancel: &CancellationToken,
stats: &Stats, stats: &Stats,
deadline: Option<Instant>,
) -> Result<()> { ) -> Result<()> {
if bytes == 0 { if bytes == 0 {
return Ok(()); return Ok(());
@@ -729,11 +739,15 @@ async fn wait_for_traffic_budget_or_cancel(
} }
let wait_started_at = Instant::now(); let wait_started_at = Instant::now();
if deadline.is_some_and(|deadline| wait_started_at >= deadline) {
stats.increment_flow_wait_middle_rate_limit_cancelled_total();
return Err(ProxyError::TrafficBudgetWaitDeadlineExceeded);
}
tokio::select! { tokio::select! {
_ = tokio::time::sleep(next_refill_delay()) => {} _ = tokio::time::sleep(next_refill_delay()) => {}
_ = cancel.cancelled() => { _ = cancel.cancelled() => {
stats.increment_flow_wait_middle_rate_limit_cancelled_total(); stats.increment_flow_wait_middle_rate_limit_cancelled_total();
return Err(ProxyError::Proxy("traffic budget wait cancelled".into())); return Err(ProxyError::TrafficBudgetWaitCancelled);
} }
} }
let wait_ms = wait_started_at let wait_ms = wait_started_at
@@ -1188,7 +1202,7 @@ where
tokio::time::sleep(delay).await; tokio::time::sleep(delay).await;
let _ = me_pool.send_close(conn_id).await; let _ = me_pool.send_close(conn_id).await;
me_pool.registry().unregister(conn_id).await; me_pool.registry().unregister(conn_id).await;
return Err(ProxyError::Proxy(ROUTE_SWITCH_ERROR_MSG.to_string())); return Err(ProxyError::RouteSwitched);
} }
// Per-user ad_tag from access.user_ad_tags; fallback to general.ad_tag (hot-reloadable) // Per-user ad_tag from access.user_ad_tags; fallback to general.ad_tag (hot-reloadable)
@@ -1353,7 +1367,7 @@ where
} else { } else {
None None
}; };
let _ = writer.flush().await; let _ = flush_client_or_cancel(&mut writer, &flow_cancel_me_writer).await;
let flush_duration_us = flush_started_at.map(|started| { let flush_duration_us = flush_started_at.map(|started| {
started started
.elapsed() .elapsed()
@@ -1416,7 +1430,8 @@ where
} else { } else {
None None
}; };
let _ = writer.flush().await; let _ =
flush_client_or_cancel(&mut writer, &flow_cancel_me_writer).await;
let flush_duration_us = flush_started_at.map(|started| { let flush_duration_us = flush_started_at.map(|started| {
started started
.elapsed() .elapsed()
@@ -1484,7 +1499,11 @@ where
} else { } else {
None None
}; };
let _ = writer.flush().await; let _ = flush_client_or_cancel(
&mut writer,
&flow_cancel_me_writer,
)
.await;
let flush_duration_us = flush_started_at.map(|started| { let flush_duration_us = flush_started_at.map(|started| {
started started
.elapsed() .elapsed()
@@ -1551,7 +1570,11 @@ where
} else { } else {
None None
}; };
let _ = writer.flush().await; let _ = flush_client_or_cancel(
&mut writer,
&flow_cancel_me_writer,
)
.await;
let flush_duration_us = flush_started_at.map(|started| { let flush_duration_us = flush_started_at.map(|started| {
started started
.elapsed() .elapsed()
@@ -1597,7 +1620,7 @@ where
} else { } else {
None None
}; };
writer.flush().await.map_err(ProxyError::Io)?; flush_client_or_cancel(&mut writer, &flow_cancel_me_writer).await?;
let flush_duration_us = flush_started_at.map(|started| { let flush_duration_us = flush_started_at.map(|started| {
started started
.elapsed() .elapsed()
@@ -1690,7 +1713,7 @@ where
stats.as_ref(), stats.as_ref(),
) )
.await; .await;
main_result = Err(ProxyError::Proxy(ROUTE_SWITCH_ERROR_MSG.to_string())); main_result = Err(ProxyError::RouteSwitched);
break; break;
} }
@@ -1721,6 +1744,7 @@ where
traffic_lease.as_ref(), traffic_lease.as_ref(),
RateDirection::Up, RateDirection::Up,
payload.len() as u64, payload.len() as u64,
None,
) )
.await?; .await?;
forensics.bytes_c2me = forensics forensics.bytes_c2me = forensics
@@ -1735,6 +1759,7 @@ where
limit, limit,
stats.as_ref(), stats.as_ref(),
&flow_cancel, &flow_cancel,
None,
) )
.await .await
{ {
@@ -1757,6 +1782,12 @@ where
)); ));
break; break;
} }
Err(MiddleQuotaReserveError::DeadlineExceeded) => {
main_result = Err(ProxyError::Proxy(
"ME C->ME quota reservation deadline exceeded".into(),
));
break;
}
} }
stats.add_user_octets_from_handle(user_stats, payload.len() as u64); stats.add_user_octets_from_handle(user_stats, payload.len() as u64);
} else { } else {
@@ -1852,9 +1883,7 @@ where
}; };
// When client closes, but ME channel stopped as unregistered - it isnt error // When client closes, but ME channel stopped as unregistered - it isnt error
if client_closed if client_closed && matches!(writer_result, Err(ProxyError::MiddleConnectionLost)) {
&& matches!(writer_result, Err(ProxyError::MiddleConnectionLost))
{
writer_result = Ok(()); writer_result = Ok(());
} }
@@ -2450,8 +2479,10 @@ where
let data_len = data.len() as u64; let data_len = data.len() as u64;
if let (Some(limit), Some(user_stats)) = (quota_limit, quota_user_stats) { if let (Some(limit), Some(user_stats)) = (quota_limit, quota_user_stats) {
let soft_limit = quota_soft_cap(limit, quota_soft_overshoot_bytes); let soft_limit = quota_soft_cap(limit, quota_soft_overshoot_bytes);
match reserve_user_quota_with_yield(user_stats, data_len, soft_limit, stats, cancel) match reserve_user_quota_with_yield(
.await user_stats, data_len, soft_limit, stats, cancel, None,
)
.await
{ {
Ok(_) => {} Ok(_) => {}
Err(MiddleQuotaReserveError::LimitExceeded) => { Err(MiddleQuotaReserveError::LimitExceeded) => {
@@ -2470,6 +2501,11 @@ where
"ME D->C quota reservation cancelled".into(), "ME D->C quota reservation cancelled".into(),
)); ));
} }
Err(MiddleQuotaReserveError::DeadlineExceeded) => {
return Err(ProxyError::Proxy(
"ME D->C quota reservation deadline exceeded".into(),
));
}
} }
} }
wait_for_traffic_budget_or_cancel( wait_for_traffic_budget_or_cancel(
@@ -2478,22 +2514,30 @@ where
data_len, data_len,
cancel, cancel,
stats, stats,
None,
) )
.await?; .await?;
let write_mode = let write_mode = match write_client_payload(
match write_client_payload(client_writer, proto_tag, flags, &data, rng, frame_buf) client_writer,
.await proto_tag,
{ flags,
Ok(mode) => mode, &data,
Err(err) => { rng,
if quota_limit.is_some() { frame_buf,
stats.add_quota_write_fail_bytes_total(data_len); cancel,
stats.increment_quota_write_fail_events_total(); )
} .await
return Err(err); {
Ok(mode) => mode,
Err(err) => {
if quota_limit.is_some() {
stats.add_quota_write_fail_bytes_total(data_len);
stats.increment_quota_write_fail_events_total();
} }
}; return Err(err);
}
};
bytes_me2c.fetch_add(data_len, Ordering::Relaxed); bytes_me2c.fetch_add(data_len, Ordering::Relaxed);
if let Some(user_stats) = quota_user_stats { if let Some(user_stats) = quota_user_stats {
@@ -2517,9 +2561,16 @@ where
} else { } else {
trace!(conn_id, confirm, "ME->C quickack"); trace!(conn_id, confirm, "ME->C quickack");
} }
wait_for_traffic_budget_or_cancel(traffic_lease, RateDirection::Down, 4, cancel, stats) wait_for_traffic_budget_or_cancel(
.await?; traffic_lease,
write_client_ack(client_writer, proto_tag, confirm).await?; RateDirection::Down,
4,
cancel,
stats,
None,
)
.await?;
write_client_ack(client_writer, proto_tag, confirm, cancel).await?;
stats.increment_me_d2c_ack_frames_total(); stats.increment_me_d2c_ack_frames_total();
Ok(MeWriterResponseOutcome::Continue { Ok(MeWriterResponseOutcome::Continue {
@@ -2571,6 +2622,7 @@ async fn write_client_payload<W>(
data: &[u8], data: &[u8],
rng: &SecureRandom, rng: &SecureRandom,
frame_buf: &mut Vec<u8>, frame_buf: &mut Vec<u8>,
cancel: &CancellationToken,
) -> Result<MeD2cWriteMode> ) -> Result<MeD2cWriteMode>
where where
W: AsyncWrite + Unpin + Send + 'static, W: AsyncWrite + Unpin + Send + 'static,
@@ -2598,21 +2650,12 @@ where
frame_buf.reserve(wire_len); frame_buf.reserve(wire_len);
frame_buf.push(first); frame_buf.push(first);
frame_buf.extend_from_slice(data); frame_buf.extend_from_slice(data);
client_writer write_all_client_or_cancel(client_writer, frame_buf.as_slice(), cancel).await?;
.write_all(frame_buf.as_slice())
.await
.map_err(ProxyError::Io)?;
MeD2cWriteMode::Coalesced MeD2cWriteMode::Coalesced
} else { } else {
let header = [first]; let header = [first];
client_writer write_all_client_or_cancel(client_writer, &header, cancel).await?;
.write_all(&header) write_all_client_or_cancel(client_writer, data, cancel).await?;
.await
.map_err(ProxyError::Io)?;
client_writer
.write_all(data)
.await
.map_err(ProxyError::Io)?;
MeD2cWriteMode::Split MeD2cWriteMode::Split
} }
} else if len_words < (1 << 24) { } else if len_words < (1 << 24) {
@@ -2627,21 +2670,12 @@ where
frame_buf.reserve(wire_len); frame_buf.reserve(wire_len);
frame_buf.extend_from_slice(&[first, lw[0], lw[1], lw[2]]); frame_buf.extend_from_slice(&[first, lw[0], lw[1], lw[2]]);
frame_buf.extend_from_slice(data); frame_buf.extend_from_slice(data);
client_writer write_all_client_or_cancel(client_writer, frame_buf.as_slice(), cancel).await?;
.write_all(frame_buf.as_slice())
.await
.map_err(ProxyError::Io)?;
MeD2cWriteMode::Coalesced MeD2cWriteMode::Coalesced
} else { } else {
let header = [first, lw[0], lw[1], lw[2]]; let header = [first, lw[0], lw[1], lw[2]];
client_writer write_all_client_or_cancel(client_writer, &header, cancel).await?;
.write_all(&header) write_all_client_or_cancel(client_writer, data, cancel).await?;
.await
.map_err(ProxyError::Io)?;
client_writer
.write_all(data)
.await
.map_err(ProxyError::Io)?;
MeD2cWriteMode::Split MeD2cWriteMode::Split
} }
} else { } else {
@@ -2676,21 +2710,12 @@ where
frame_buf.resize(start + padding_len, 0); frame_buf.resize(start + padding_len, 0);
rng.fill(&mut frame_buf[start..]); rng.fill(&mut frame_buf[start..]);
} }
client_writer write_all_client_or_cancel(client_writer, frame_buf.as_slice(), cancel).await?;
.write_all(frame_buf.as_slice())
.await
.map_err(ProxyError::Io)?;
MeD2cWriteMode::Coalesced MeD2cWriteMode::Coalesced
} else { } else {
let header = len_val.to_le_bytes(); let header = len_val.to_le_bytes();
client_writer write_all_client_or_cancel(client_writer, &header, cancel).await?;
.write_all(&header) write_all_client_or_cancel(client_writer, data, cancel).await?;
.await
.map_err(ProxyError::Io)?;
client_writer
.write_all(data)
.await
.map_err(ProxyError::Io)?;
if padding_len > 0 { if padding_len > 0 {
frame_buf.clear(); frame_buf.clear();
if frame_buf.capacity() < padding_len { if frame_buf.capacity() < padding_len {
@@ -2698,10 +2723,7 @@ where
} }
frame_buf.resize(padding_len, 0); frame_buf.resize(padding_len, 0);
rng.fill(frame_buf.as_mut_slice()); rng.fill(frame_buf.as_mut_slice());
client_writer write_all_client_or_cancel(client_writer, frame_buf.as_slice(), cancel).await?;
.write_all(frame_buf.as_slice())
.await
.map_err(ProxyError::Io)?;
} }
MeD2cWriteMode::Split MeD2cWriteMode::Split
} }
@@ -2715,6 +2737,7 @@ async fn write_client_ack<W>(
client_writer: &mut CryptoWriter<W>, client_writer: &mut CryptoWriter<W>,
proto_tag: ProtoTag, proto_tag: ProtoTag,
confirm: u32, confirm: u32,
cancel: &CancellationToken,
) -> Result<()> ) -> Result<()>
where where
W: AsyncWrite + Unpin + Send + 'static, W: AsyncWrite + Unpin + Send + 'static,
@@ -2724,10 +2747,34 @@ where
} else { } else {
confirm.to_le_bytes() confirm.to_le_bytes()
}; };
client_writer write_all_client_or_cancel(client_writer, &bytes, cancel).await
.write_all(&bytes) }
.await
.map_err(ProxyError::Io) async fn write_all_client_or_cancel<W>(
client_writer: &mut CryptoWriter<W>,
bytes: &[u8],
cancel: &CancellationToken,
) -> Result<()>
where
W: AsyncWrite + Unpin + Send + 'static,
{
tokio::select! {
result = client_writer.write_all(bytes) => result.map_err(ProxyError::Io),
_ = cancel.cancelled() => Err(ProxyError::MiddleClientWriterCancelled),
}
}
async fn flush_client_or_cancel<W>(
client_writer: &mut CryptoWriter<W>,
cancel: &CancellationToken,
) -> Result<()>
where
W: AsyncWrite + Unpin + Send + 'static,
{
tokio::select! {
result = client_writer.flush() => result.map_err(ProxyError::Io),
_ = cancel.cancelled() => Err(ProxyError::MiddleClientWriterCancelled),
}
} }
#[cfg(test)] #[cfg(test)]
+2 -9
View File
@@ -518,10 +518,7 @@ impl<S: AsyncRead + Unpin> AsyncRead for StatsIo<S> {
Poll::Ready(Ok(n)) => { Poll::Ready(Ok(n)) => {
if reserved_read_bytes > n as u64 { if reserved_read_bytes > n as u64 {
let refund_bytes = reserved_read_bytes - n as u64; let refund_bytes = reserved_read_bytes - n as u64;
refund_reserved_quota_bytes( refund_reserved_quota_bytes(this.user_stats.as_ref(), refund_bytes);
this.user_stats.as_ref(),
refund_bytes,
);
this.stats.add_quota_refund_bytes_total(refund_bytes); this.stats.add_quota_refund_bytes_total(refund_bytes);
} }
if n > 0 { if n > 0 {
@@ -538,7 +535,6 @@ impl<S: AsyncRead + Unpin> AsyncRead for StatsIo<S> {
this.quota_bytes_since_check = 0; this.quota_bytes_since_check = 0;
} }
} }
} }
if let Some(limit) = this.quota_limit if let Some(limit) = this.quota_limit
&& this.user_stats.quota_used() >= limit && this.user_stats.quota_used() >= limit
@@ -700,10 +696,7 @@ impl<S: AsyncWrite + Unpin> AsyncWrite for StatsIo<S> {
Poll::Ready(Ok(n)) => { Poll::Ready(Ok(n)) => {
if reserved_bytes > n as u64 { if reserved_bytes > n as u64 {
let refund_bytes = reserved_bytes - n as u64; let refund_bytes = reserved_bytes - n as u64;
refund_reserved_quota_bytes( refund_reserved_quota_bytes(this.user_stats.as_ref(), refund_bytes);
this.user_stats.as_ref(),
refund_bytes,
);
this.stats.add_quota_refund_bytes_total(refund_bytes); this.stats.add_quota_refund_bytes_total(refund_bytes);
} }
if shaper_reserved_bytes > n as u64 if shaper_reserved_bytes > n as u64
-2
View File
@@ -4,8 +4,6 @@ use std::time::{Duration, SystemTime, UNIX_EPOCH};
use tokio::sync::watch; use tokio::sync::watch;
pub(crate) const ROUTE_SWITCH_ERROR_MSG: &str = "Session terminated";
#[derive(Clone, Copy, Debug, PartialEq, Eq)] #[derive(Clone, Copy, Debug, PartialEq, Eq)]
#[repr(u8)] #[repr(u8)]
pub(crate) enum RelayRouteMode { pub(crate) enum RelayRouteMode {
+1 -1
View File
@@ -661,7 +661,7 @@ async fn integration_route_cutover_and_quota_overlap_fails_closed_and_releases_s
assert!( assert!(
matches!(relay_result, Err(ProxyError::DataQuotaExceeded { .. })) matches!(relay_result, Err(ProxyError::DataQuotaExceeded { .. }))
|| matches!(relay_result, Err(ProxyError::Proxy(ref msg)) if msg == crate::proxy::route_mode::ROUTE_SWITCH_ERROR_MSG), || matches!(relay_result, Err(ProxyError::RouteSwitched)),
"overlap race must fail closed via quota enforcement or generic cutover termination" "overlap race must fail closed via quota enforcement or generic cutover termination"
); );
+37 -16
View File
@@ -637,6 +637,22 @@ fn unknown_dc_log_path_revalidation_rejects_parent_swapped_to_symlink() {
"telemt-unknown-dc-parent-swap-{}", "telemt-unknown-dc-parent-swap-{}",
std::process::id() std::process::id()
)); ));
if let Ok(meta) = fs::symlink_metadata(&parent) {
if meta.file_type().is_symlink() || meta.is_file() {
fs::remove_file(&parent).expect("stale parent-swap path must be removable");
} else {
fs::remove_dir_all(&parent).expect("stale parent-swap directory must be removable");
}
}
let moved = parent.with_extension("bak");
if let Ok(meta) = fs::symlink_metadata(&moved) {
if meta.file_type().is_symlink() || meta.is_file() {
fs::remove_file(&moved).expect("stale parent-swap backup path must be removable");
} else {
fs::remove_dir_all(&moved)
.expect("stale parent-swap backup directory must be removable");
}
}
fs::create_dir_all(&parent).expect("parent-swap test parent must be creatable"); fs::create_dir_all(&parent).expect("parent-swap test parent must be creatable");
let rel_candidate = format!( let rel_candidate = format!(
@@ -646,8 +662,6 @@ fn unknown_dc_log_path_revalidation_rejects_parent_swapped_to_symlink() {
let sanitized = sanitize_unknown_dc_log_path(&rel_candidate) let sanitized = sanitize_unknown_dc_log_path(&rel_candidate)
.expect("candidate must sanitize before parent swap"); .expect("candidate must sanitize before parent swap");
let moved = parent.with_extension("bak");
let _ = fs::remove_dir_all(&moved);
fs::rename(&parent, &moved).expect("parent must be movable for swap simulation"); fs::rename(&parent, &moved).expect("parent must be movable for swap simulation");
symlink("/tmp", &parent).expect("symlink replacement for parent must be creatable"); symlink("/tmp", &parent).expect("symlink replacement for parent must be creatable");
@@ -720,6 +734,24 @@ fn adversarial_parent_swap_after_check_is_blocked_by_anchored_open() {
"telemt-unknown-dc-parent-swap-openat-{}", "telemt-unknown-dc-parent-swap-openat-{}",
std::process::id() std::process::id()
)); ));
if let Ok(meta) = fs::symlink_metadata(&base) {
if meta.file_type().is_symlink() || meta.is_file() {
fs::remove_file(&base).expect("stale parent-swap-openat path must be removable");
} else {
fs::remove_dir_all(&base)
.expect("stale parent-swap-openat directory must be removable");
}
}
let moved = base.with_extension("bak");
if let Ok(meta) = fs::symlink_metadata(&moved) {
if meta.file_type().is_symlink() || meta.is_file() {
fs::remove_file(&moved)
.expect("stale parent-swap-openat backup path must be removable");
} else {
fs::remove_dir_all(&moved)
.expect("stale parent-swap-openat backup directory must be removable");
}
}
fs::create_dir_all(&base).expect("parent-swap-openat base must be creatable"); fs::create_dir_all(&base).expect("parent-swap-openat base must be creatable");
let rel_candidate = format!( let rel_candidate = format!(
@@ -743,8 +775,6 @@ fn adversarial_parent_swap_after_check_is_blocked_by_anchored_open() {
let outside_target = outside_parent.join("unknown-dc.log"); let outside_target = outside_parent.join("unknown-dc.log");
let _ = fs::remove_file(&outside_target); let _ = fs::remove_file(&outside_target);
let moved = base.with_extension("bak");
let _ = fs::remove_dir_all(&moved);
fs::rename(&base, &moved).expect("base parent must be movable for swap simulation"); fs::rename(&base, &moved).expect("base parent must be movable for swap simulation");
symlink(&outside_parent, &base).expect("base parent symlink replacement must be creatable"); symlink(&outside_parent, &base).expect("base parent symlink replacement must be creatable");
@@ -1489,10 +1519,7 @@ async fn direct_relay_cutover_midflight_releases_route_gauge() {
"cutover should terminate direct relay session" "cutover should terminate direct relay session"
); );
assert!( assert!(
matches!( matches!(relay_result, Err(ProxyError::RouteSwitched)),
relay_result,
Err(ProxyError::Proxy(ref msg)) if msg == ROUTE_SWITCH_ERROR_MSG
),
"client-visible cutover error must stay generic and avoid route-internal metadata" "client-visible cutover error must stay generic and avoid route-internal metadata"
); );
@@ -1629,10 +1656,7 @@ async fn direct_relay_cutover_storm_multi_session_keeps_generic_errors_and_relea
.expect("direct relay task must not panic"); .expect("direct relay task must not panic");
assert!( assert!(
matches!( matches!(relay_result, Err(ProxyError::RouteSwitched)),
relay_result,
Err(ProxyError::Proxy(ref msg)) if msg == ROUTE_SWITCH_ERROR_MSG
),
"storm-cutover termination must remain generic for all direct sessions" "storm-cutover termination must remain generic for all direct sessions"
); );
} }
@@ -1935,10 +1959,7 @@ async fn adversarial_direct_relay_cutover_integrity() {
.expect("Session must not panic"); .expect("Session must not panic");
assert!( assert!(
matches!( matches!(result, Err(ProxyError::RouteSwitched)),
result,
Err(ProxyError::Proxy(ref msg)) if msg == ROUTE_SWITCH_ERROR_MSG
),
"Session must terminate with route switch error on cutover" "Session must terminate with route switch error on cutover"
); );
} }
@@ -13,6 +13,8 @@ struct CountedWriter {
fail_writes: bool, fail_writes: bool,
} }
struct StalledWriter;
impl CountedWriter { impl CountedWriter {
fn new(write_calls: Arc<AtomicUsize>, fail_writes: bool) -> Self { fn new(write_calls: Arc<AtomicUsize>, fail_writes: bool) -> Self {
Self { Self {
@@ -49,12 +51,36 @@ impl AsyncWrite for CountedWriter {
} }
} }
impl AsyncWrite for StalledWriter {
fn poll_write(
self: Pin<&mut Self>,
_cx: &mut Context<'_>,
_buf: &[u8],
) -> Poll<io::Result<usize>> {
Poll::Pending
}
fn poll_flush(self: Pin<&mut Self>, _cx: &mut Context<'_>) -> Poll<io::Result<()>> {
Poll::Pending
}
fn poll_shutdown(self: Pin<&mut Self>, _cx: &mut Context<'_>) -> Poll<io::Result<()>> {
Poll::Pending
}
}
fn make_crypto_writer(inner: CountedWriter) -> CryptoWriter<CountedWriter> { fn make_crypto_writer(inner: CountedWriter) -> CryptoWriter<CountedWriter> {
let key = [0u8; 32]; let key = [0u8; 32];
let iv = 0u128; let iv = 0u128;
CryptoWriter::new(inner, AesCtr::new(&key, iv), 8 * 1024) CryptoWriter::new(inner, AesCtr::new(&key, iv), 8 * 1024)
} }
fn make_stalled_crypto_writer() -> CryptoWriter<StalledWriter> {
let key = [0u8; 32];
let iv = 0u128;
CryptoWriter::new(StalledWriter, AesCtr::new(&key, iv), 8 * 1024)
}
#[tokio::test] #[tokio::test]
async fn me_writer_write_fail_keeps_reserved_quota_and_tracks_fail_metrics() { async fn me_writer_write_fail_keeps_reserved_quota_and_tracks_fail_metrics() {
let stats = Stats::new(); let stats = Stats::new();
@@ -189,3 +215,53 @@ async fn me_writer_pre_write_quota_reject_happens_before_writer_poll() {
); );
assert_eq!(bytes_me2c.load(Ordering::Relaxed), 0); assert_eq!(bytes_me2c.load(Ordering::Relaxed), 0);
} }
#[tokio::test]
async fn me_writer_data_write_obeys_flow_cancellation() {
let stats = Stats::new();
let user = "middle-me-writer-cancel-user";
let mut writer = make_stalled_crypto_writer();
let mut frame_buf = Vec::new();
let bytes_me2c = AtomicU64::new(0);
let cancel = CancellationToken::new();
cancel.cancel();
let result = process_me_writer_response_with_traffic_lease(
MeResponse::Data {
flags: 0,
data: Bytes::from_static(&[0x31, 0x32, 0x33, 0x34]),
route_permit: None,
},
&mut writer,
ProtoTag::Intermediate,
&SecureRandom::new(),
&mut frame_buf,
&stats,
user,
None,
None,
0,
None,
&cancel,
&bytes_me2c,
13,
true,
false,
)
.await;
assert!(
matches!(result, Err(ProxyError::MiddleClientWriterCancelled)),
"cancelled middle writer must return a bounded cancellation error"
);
assert_eq!(
bytes_me2c.load(Ordering::Relaxed),
0,
"cancelled write must not advance committed ME->C bytes"
);
assert_eq!(
stats.get_user_total_octets(user),
0,
"cancelled write must not advance user output telemetry"
);
}
@@ -4,10 +4,67 @@ use std::io;
use std::pin::Pin; use std::pin::Pin;
use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering}; use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering};
use std::sync::{Arc, Mutex}; use std::sync::{Arc, Mutex};
use std::task::{Context, Poll}; use std::task::{Context, Poll, Wake};
use tokio::io::{AsyncWrite, AsyncWriteExt}; use tokio::io::{AsyncRead, AsyncWrite, AsyncWriteExt, ReadBuf};
use tokio::time::Instant; use tokio::time::Instant;
enum ReadStep {
Data(Vec<u8>),
Pending,
Eof,
Error,
}
struct ScriptedReader {
scripted_reads: Arc<Mutex<VecDeque<ReadStep>>>,
read_calls: Arc<AtomicUsize>,
}
impl ScriptedReader {
fn new(script: Vec<ReadStep>, read_calls: Arc<AtomicUsize>) -> Self {
Self {
scripted_reads: Arc::new(Mutex::new(script.into())),
read_calls,
}
}
}
impl AsyncRead for ScriptedReader {
fn poll_read(
self: Pin<&mut Self>,
_cx: &mut Context<'_>,
buf: &mut ReadBuf<'_>,
) -> Poll<io::Result<()>> {
let this = self.get_mut();
this.read_calls.fetch_add(1, Ordering::Relaxed);
let step = this
.scripted_reads
.lock()
.unwrap_or_else(|poisoned| poisoned.into_inner())
.pop_front()
.unwrap_or(ReadStep::Eof);
match step {
ReadStep::Data(data) => {
let n = data.len().min(buf.remaining());
buf.put_slice(&data[..n]);
Poll::Ready(Ok(()))
}
ReadStep::Pending => Poll::Pending,
ReadStep::Eof => Poll::Ready(Ok(())),
ReadStep::Error => Poll::Ready(Err(io::Error::new(
io::ErrorKind::BrokenPipe,
"forced read failure",
))),
}
}
}
struct NoopWake;
impl Wake for NoopWake {
fn wake(self: Arc<Self>) {}
}
struct ScriptedWriter { struct ScriptedWriter {
scripted_writes: Arc<Mutex<VecDeque<usize>>>, scripted_writes: Arc<Mutex<VecDeque<usize>>>,
write_calls: Arc<AtomicUsize>, write_calls: Arc<AtomicUsize>,
@@ -80,6 +137,127 @@ fn make_stats_io_with_script(
(io, stats, write_calls, quota_exceeded) (io, stats, write_calls, quota_exceeded)
} }
fn make_stats_io_with_read_script(
user: &str,
quota_limit: u64,
precharged_quota: u64,
script: Vec<ReadStep>,
) -> (
StatsIo<ScriptedReader>,
Arc<Stats>,
Arc<AtomicUsize>,
Arc<AtomicBool>,
) {
let stats = Arc::new(Stats::new());
if precharged_quota > 0 {
let user_stats = stats.get_or_create_user_stats_handle(user);
stats.quota_charge_post_write(user_stats.as_ref(), precharged_quota);
}
let read_calls = Arc::new(AtomicUsize::new(0));
let quota_exceeded = Arc::new(AtomicBool::new(false));
let io = StatsIo::new(
ScriptedReader::new(script, read_calls.clone()),
Arc::new(SharedCounters::new()),
stats.clone(),
user.to_string(),
Some(quota_limit),
quota_exceeded.clone(),
Instant::now(),
);
(io, stats, read_calls, quota_exceeded)
}
fn poll_read_once<R: AsyncRead + Unpin>(
io: &mut StatsIo<R>,
storage: &mut [u8],
) -> Poll<io::Result<usize>> {
let waker = Arc::new(NoopWake).into();
let mut cx = Context::from_waker(&waker);
let mut read_buf = ReadBuf::new(storage);
let before = read_buf.filled().len();
match Pin::new(io).poll_read(&mut cx, &mut read_buf) {
Poll::Ready(Ok(())) => Poll::Ready(Ok(read_buf.filled().len() - before)),
Poll::Ready(Err(error)) => Poll::Ready(Err(error)),
Poll::Pending => Poll::Pending,
}
}
#[test]
fn direct_c2s_quota_refunds_unused_on_short_read() {
let user = "direct-c2s-short-read-refund-user";
let (mut io, stats, read_calls, quota_exceeded) =
make_stats_io_with_read_script(user, 64, 0, vec![ReadStep::Data(vec![0x11; 5])]);
let mut storage = [0u8; 16];
let n = match poll_read_once(&mut io, &mut storage) {
Poll::Ready(Ok(n)) => n,
other => panic!("short read must complete, got {other:?}"),
};
assert_eq!(n, 5);
assert_eq!(read_calls.load(Ordering::Relaxed), 1);
assert_eq!(stats.get_user_quota_used(user), 5);
assert_eq!(stats.get_quota_refund_bytes_total(), 11);
assert!(!quota_exceeded.load(Ordering::Acquire));
}
#[test]
fn direct_c2s_quota_refunds_full_reservation_on_pending() {
let user = "direct-c2s-pending-refund-user";
let (mut io, stats, read_calls, quota_exceeded) =
make_stats_io_with_read_script(user, 64, 0, vec![ReadStep::Pending]);
let mut storage = [0u8; 16];
assert!(matches!(
poll_read_once(&mut io, &mut storage),
Poll::Pending
));
assert_eq!(read_calls.load(Ordering::Relaxed), 1);
assert_eq!(stats.get_user_quota_used(user), 0);
assert_eq!(stats.get_quota_refund_bytes_total(), 16);
assert!(!quota_exceeded.load(Ordering::Acquire));
}
#[test]
fn direct_c2s_quota_refunds_full_reservation_on_eof() {
let user = "direct-c2s-eof-refund-user";
let (mut io, stats, read_calls, quota_exceeded) =
make_stats_io_with_read_script(user, 64, 0, vec![ReadStep::Eof]);
let mut storage = [0u8; 16];
let n = match poll_read_once(&mut io, &mut storage) {
Poll::Ready(Ok(n)) => n,
other => panic!("EOF read must complete with zero bytes, got {other:?}"),
};
assert_eq!(n, 0);
assert_eq!(read_calls.load(Ordering::Relaxed), 1);
assert_eq!(stats.get_user_quota_used(user), 0);
assert_eq!(stats.get_quota_refund_bytes_total(), 16);
assert!(!quota_exceeded.load(Ordering::Acquire));
}
#[test]
fn direct_c2s_quota_refunds_full_reservation_on_error() {
let user = "direct-c2s-error-refund-user";
let (mut io, stats, read_calls, quota_exceeded) =
make_stats_io_with_read_script(user, 64, 0, vec![ReadStep::Error]);
let mut storage = [0u8; 16];
let error = match poll_read_once(&mut io, &mut storage) {
Poll::Ready(Err(error)) => error,
other => panic!("error read must return error, got {other:?}"),
};
assert_eq!(error.kind(), io::ErrorKind::BrokenPipe);
assert_eq!(read_calls.load(Ordering::Relaxed), 1);
assert_eq!(stats.get_user_quota_used(user), 0);
assert_eq!(stats.get_quota_refund_bytes_total(), 16);
assert!(!quota_exceeded.load(Ordering::Acquire));
}
#[tokio::test] #[tokio::test]
async fn direct_partial_write_charges_only_committed_bytes_without_double_charge() { async fn direct_partial_write_charges_only_committed_bytes_without_double_charge() {
let user = "direct-partial-charge-user"; let user = "direct-partial-charge-user";
+3 -9
View File
@@ -8,8 +8,8 @@ pub mod telemetry;
use dashmap::DashMap; use dashmap::DashMap;
use lru::LruCache; use lru::LruCache;
use parking_lot::Mutex; use parking_lot::Mutex;
use std::collections::{HashMap, VecDeque};
use std::collections::hash_map::DefaultHasher; use std::collections::hash_map::DefaultHasher;
use std::collections::{HashMap, VecDeque};
use std::hash::{Hash, Hasher}; use std::hash::{Hash, Hasher};
use std::num::NonZeroUsize; use std::num::NonZeroUsize;
use std::sync::Arc; use std::sync::Arc;
@@ -2378,8 +2378,7 @@ impl Stats {
self.quota_contention_total.load(Ordering::Relaxed) self.quota_contention_total.load(Ordering::Relaxed)
} }
pub fn get_quota_contention_timeout_total(&self) -> u64 { pub fn get_quota_contention_timeout_total(&self) -> u64 {
self.quota_contention_timeout_total self.quota_contention_timeout_total.load(Ordering::Relaxed)
.load(Ordering::Relaxed)
} }
pub fn get_quota_acquire_cancelled_total(&self) -> u64 { pub fn get_quota_acquire_cancelled_total(&self) -> u64 {
self.quota_acquire_cancelled_total.load(Ordering::Relaxed) self.quota_acquire_cancelled_total.load(Ordering::Relaxed)
@@ -2536,12 +2535,7 @@ impl Stats {
.unwrap_or(0) .unwrap_or(0)
} }
pub fn load_user_quota_state( pub fn load_user_quota_state(&self, user: &str, used_bytes: u64, last_reset_epoch_secs: u64) {
&self,
user: &str,
used_bytes: u64,
last_reset_epoch_secs: u64,
) {
let stats = self.get_or_create_user_stats_handle(user); let stats = self.get_or_create_user_stats_handle(user);
stats.quota_used.store(used_bytes, Ordering::Relaxed); stats.quota_used.store(used_bytes, Ordering::Relaxed);
stats stats
+4 -3
View File
@@ -150,9 +150,10 @@ impl TlsFrontCache {
is_default: cached.domain == "default", is_default: cached.domain == "default",
has_cert_info: cached.cert_info.is_some(), has_cert_info: cached.cert_info.is_some(),
has_cert_payload: cached.cert_payload.is_some(), has_cert_payload: cached.cert_payload.is_some(),
app_data_records: cached.app_data_records_sizes.len().max( app_data_records: cached
behavior.app_data_record_sizes.len(), .app_data_records_sizes
), .len()
.max(behavior.app_data_record_sizes.len()),
ticket_records: behavior.ticket_record_sizes.len(), ticket_records: behavior.ticket_record_sizes.len(),
change_cipher_spec_count: behavior.change_cipher_spec_count, change_cipher_spec_count: behavior.change_cipher_spec_count,
total_app_data_len: cached.total_app_data_len, total_app_data_len: cached.total_app_data_len,