This commit is contained in:
Mateusz Gruszczyński committed 2026-10-03 18:10:20 +02:00
1 parent 1297ea2920
commit dc0fcabf51
49 files changed
+2632 -252

No files matched your search

+445 -4
View File
@@ -16,20 +16,23 @@ use serde::{Deserialize, Serialize};
use serde_json::{json, Map, Value};
use std::{
collections::{BTreeMap, VecDeque},
io::{self, Write as IoWrite},
path::PathBuf,
sync::{
atomic::{AtomicU64, Ordering},
Arc, Mutex,
Arc, Mutex, OnceLock,
},
time::{Duration, SystemTime, UNIX_EPOCH},
time::{Duration, Instant, SystemTime, UNIX_EPOCH},
};
use tokio::{
fs::{File, OpenOptions},
io::AsyncWriteExt,
net::TcpStream,
sync::{broadcast, mpsc},
sync::{broadcast, mpsc, watch},
};
use tokio_util::sync::CancellationToken;
use tracing::{Event, Level, Subscriber};
use tracing_subscriber::{layer::Context as LayerContext, Layer};
pub const LIVE_WINDOW: usize = 256;
pub const HISTORY_CAPACITY: usize = 4096;
@@ -238,6 +241,411 @@ fn source_value_name(name: &str) -> bool {
)
}
#[derive(Debug, Clone, Copy, Serialize, Deserialize, PartialEq, Eq)]
pub enum ErrorLogFormat {
#[serde(rename = "JSON")]
Json,
#[serde(rename = "RAW")]
Raw,
}
impl Default for ErrorLogFormat {
fn default() -> Self { Self::Json }
}
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(default, deny_unknown_fields)]
pub struct ErrorLogSettings {
pub enabled: bool,
pub path: String,
pub format: ErrorLogFormat,
}
impl Default for ErrorLogSettings {
fn default() -> Self {
Self { enabled: false, path: "/var/log/rustproxy/error.log".into(), format: ErrorLogFormat::Json }
}
}
impl ErrorLogSettings {
pub fn validate(&self) -> Result<()> {
if self.enabled {
ensure!(
!self.path.trim().is_empty() && self.path.len() <= 1024 && !self.path.chars().any(char::is_control),
"error_log.path must be 1..1024 bytes without controls"
);
}
Ok(())
}
}
const ERROR_LOG_QUEUE: usize = 512;
const ERROR_LOG_RETRY: Duration = Duration::from_secs(30);
#[derive(Debug, Clone, Serialize)]
pub struct ErrorLogStats {
pub queue_capacity: usize,
pub queue_dropped: u64,
pub written: u64,
pub errors: u64,
}
#[derive(Debug)]
struct ErrorEvent {
timestamp: String,
target: String,
message: String,
fields: BTreeMap<String, String>,
}
pub struct ErrorLogController {
settings: Arc<ArcSwap<ErrorLogSettings>>,
settings_changed: watch::Sender<()>,
sender: mpsc::Sender<ErrorEvent>,
dropped: Arc<AtomicU64>,
written: Arc<AtomicU64>,
errors: Arc<AtomicU64>,
}
impl ErrorLogController {
pub fn start() -> Arc<Self> {
let (sender, receiver) = mpsc::channel(ERROR_LOG_QUEUE);
let (settings_changed, settings_watch) = watch::channel(());
let settings = Arc::new(ArcSwap::from_pointee(ErrorLogSettings::default()));
let dropped = Arc::new(AtomicU64::new(0));
let written = Arc::new(AtomicU64::new(0));
let errors = Arc::new(AtomicU64::new(0));
tokio::spawn(error_log_loop(
receiver,
settings.clone(),
settings_watch,
written.clone(),
errors.clone(),
));
Arc::new(Self { settings, settings_changed, sender, dropped, written, errors })
}
pub fn configure(&self, settings: ErrorLogSettings) {
self.settings.store(Arc::new(settings));
self.settings_changed.send_replace(());
}
fn push(&self, event: ErrorEvent) {
if self.settings.load().enabled && self.sender.try_send(event).is_err() {
self.dropped.fetch_add(1, Ordering::Relaxed);
}
}
pub fn stats(&self) -> ErrorLogStats {
ErrorLogStats {
queue_capacity: ERROR_LOG_QUEUE,
queue_dropped: self.dropped.load(Ordering::Relaxed),
written: self.written.load(Ordering::Relaxed),
errors: self.errors.load(Ordering::Relaxed),
}
}
}
static ERROR_LOG: OnceLock<Arc<ErrorLogController>> = OnceLock::new();
pub fn install_error_log_controller(controller: Arc<ErrorLogController>) -> Result<()> {
ERROR_LOG.set(controller).map_err(|_| anyhow::anyhow!("error logger already initialized"))
}
pub fn configure_error_log(settings: ErrorLogSettings) {
if let Some(controller) = ERROR_LOG.get() { controller.configure(settings); }
}
pub fn error_log_stats() -> Option<ErrorLogStats> {
ERROR_LOG.get().map(|controller| controller.stats())
}
#[derive(Clone)]
pub struct ErrorLogLayer {
controller: Arc<ErrorLogController>,
}
impl ErrorLogLayer {
pub fn new(controller: Arc<ErrorLogController>) -> Self { Self { controller } }
}
#[derive(Default)]
struct EventVisitor {
message: Option<String>,
fields: BTreeMap<String, String>,
}
impl tracing::field::Visit for EventVisitor {
fn record_debug(&mut self, field: &tracing::field::Field, value: &dyn std::fmt::Debug) {
let value = format!("{value:?}");
if field.name() == "message" { self.message = Some(value); }
else { self.fields.insert(field.name().to_owned(), value); }
}
fn record_str(&mut self, field: &tracing::field::Field, value: &str) {
if field.name() == "message" { self.message = Some(value.to_owned()); }
else { self.fields.insert(field.name().to_owned(), value.to_owned()); }
}
fn record_i64(&mut self, field: &tracing::field::Field, value: i64) { self.fields.insert(field.name().to_owned(), value.to_string()); }
fn record_u64(&mut self, field: &tracing::field::Field, value: u64) { self.fields.insert(field.name().to_owned(), value.to_string()); }
fn record_bool(&mut self, field: &tracing::field::Field, value: bool) { self.fields.insert(field.name().to_owned(), value.to_string()); }
fn record_f64(&mut self, field: &tracing::field::Field, value: f64) { self.fields.insert(field.name().to_owned(), value.to_string()); }
}
impl<S: Subscriber> Layer<S> for ErrorLogLayer {
fn on_event(&self, event: &Event<'_>, _ctx: LayerContext<'_, S>) {
if *event.metadata().level() != Level::ERROR { return; }
let mut visitor = EventVisitor::default();
event.record(&mut visitor);
let mut fields = visitor.fields;
for (key, value) in &mut fields {
*value = if is_secret_field(key) {
"[REDACTED]".into()
} else {
redact_text(value)
};
}
self.controller.push(ErrorEvent {
timestamp: rfc3339_now(),
target: event.metadata().target().to_owned(),
message: redact_text(visitor.message.as_deref().unwrap_or("error")),
fields,
});
}
}
#[derive(Clone, Copy, Default)]
pub struct RedactingStderr;
pub struct RedactingWriter { buffer: Vec<u8> }
impl IoWrite for RedactingWriter {
fn write(&mut self, bytes: &[u8]) -> io::Result<usize> { self.buffer.extend_from_slice(bytes); Ok(bytes.len()) }
fn flush(&mut self) -> io::Result<()> {
if self.buffer.is_empty() { return Ok(()); }
let text = String::from_utf8_lossy(&self.buffer);
let redacted = redact_text(&text);
let mut stderr = io::stderr().lock();
stderr.write_all(redacted.as_bytes())?;
stderr.flush()?;
self.buffer.clear();
Ok(())
}
}
impl Drop for RedactingWriter { fn drop(&mut self) { let _ = self.flush(); } }
impl<'a> tracing_subscriber::fmt::MakeWriter<'a> for RedactingStderr {
type Writer = RedactingWriter;
fn make_writer(&'a self) -> Self::Writer { RedactingWriter { buffer: Vec::with_capacity(512) } }
}
fn redact_regex(pattern: &'static str) -> &'static regex::Regex {
static AUTH: OnceLock<regex::Regex> = OnceLock::new();
static QUERY: OnceLock<regex::Regex> = OnceLock::new();
static JSON: OnceLock<regex::Regex> = OnceLock::new();
static KV: OnceLock<regex::Regex> = OnceLock::new();
static PEM: OnceLock<regex::Regex> = OnceLock::new();
match pattern {
"auth" => AUTH.get_or_init(|| regex::Regex::new(r"(?i)(authorization\s*[:=]\s*)(?:bearer\s+)?[^\s,;]+" ).expect("redaction regex")),
"query" => QUERY.get_or_init(|| regex::Regex::new(r"(?i)([?&](?:token|access_token|api_token|password|backup_password|secret|credential|credentials|key_secret|application_secret|consumer_key)=)[^&\s]+" ).expect("redaction regex")),
"json" => JSON.get_or_init(|| regex::Regex::new(r#"(?i)(\"(?:authorization|proxy_authorization|api_token|access_token|application_secret|consumer_key|key_secret|backup_password|password|credential|credentials|secret|token|private_key|master_key)\"\s*:\s*\")[^\"]*(\")"#).expect("redaction regex")),
"kv" => KV.get_or_init(|| regex::Regex::new(r"(?i)((?:authorization|proxy[_-]?authorization|api[_-]?token|access[_-]?token|application[_-]?secret|consumer[_-]?key|key[_-]?secret|backup[_-]?password|password|credential|credentials|secret|token|private[_-]?key|master[_-]?key)\s*[=:]\s*)[^\s,;\"']+" ).expect("redaction regex")),
"pem" => PEM.get_or_init(|| regex::Regex::new(r"(?s)-----BEGIN [^-\r\n]*PRIVATE KEY-----.*?-----END [^-\r\n]*PRIVATE KEY-----" ).expect("redaction regex")),
_ => unreachable!(),
}
}
fn is_secret_field(name: &str) -> bool {
matches!(
name.to_ascii_lowercase().as_str(),
"authorization"
| "proxy_authorization"
| "api_token"
| "access_token"
| "application_secret"
| "consumer_key"
| "key_secret"
| "backup_password"
| "password"
| "credential"
| "credentials"
| "secret"
| "token"
| "private_key"
| "master_key"
)
}
pub fn redact_text(input: &str) -> String {
let mut value = redact_regex("pem").replace_all(input, "[REDACTED PRIVATE KEY]").into_owned();
value = redact_regex("auth").replace_all(&value, "$1[REDACTED]").into_owned();
value = redact_regex("query").replace_all(&value, "$1[REDACTED]").into_owned();
value = redact_regex("json").replace_all(&value, "$1[REDACTED]$2").into_owned();
redact_regex("kv").replace_all(&value, "$1[REDACTED]").into_owned()
}
fn redact_value(value: &mut Value) {
match value {
Value::String(text) => *text = redact_text(text),
Value::Array(items) => for item in items { redact_value(item); },
Value::Object(items) => {
for (key, item) in items {
if is_secret_field(key) {
*item = Value::String("[REDACTED]".into());
} else {
redact_value(item);
}
}
}
_ => {}
}
}
fn rfc3339_now() -> String {
let ms = SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_millis() as i64;
let seconds = ms / 1000;
let millis = ms % 1000;
let days = seconds.div_euclid(86_400);
let day_seconds = seconds.rem_euclid(86_400);
let z = days + 719_468;
let era = if z >= 0 { z } else { z - 146_096 }.div_euclid(146_097);
let doe = z - era * 146_097;
let yoe = (doe - doe / 1_460 + doe / 36_524 - doe / 146_096) / 365;
let mut year = yoe + era * 400;
let doy = doe - (365 * yoe + yoe / 4 - yoe / 100);
let mp = (5 * doy + 2) / 153;
let day = doy - (153 * mp + 2) / 5 + 1;
let month = mp + if mp < 10 { 3 } else { -9 };
if month <= 2 { year += 1; }
let hour = day_seconds / 3600;
let minute = (day_seconds % 3600) / 60;
let second = day_seconds % 60;
format!("{year:04}-{month:02}-{day:02}T{hour:02}:{minute:02}:{second:02}.{millis:03}Z")
}
async fn open_error_file(path: PathBuf) -> Result<File> {
let file = tokio::task::spawn_blocking(move || -> Result<std::fs::File> {
let mut options = std::fs::OpenOptions::new();
options.create(true).append(true).write(true);
#[cfg(unix)]
{
use std::os::unix::fs::OpenOptionsExt;
options.mode(0o600).custom_flags(libc::O_NOFOLLOW);
}
let file = options.open(&path)?;
let metadata = file.metadata()?;
ensure!(metadata.is_file(), "error log path is not a regular file");
#[cfg(unix)]
{
use std::os::unix::fs::PermissionsExt;
file.set_permissions(std::fs::Permissions::from_mode(0o600))?;
}
Ok(file)
}).await.map_err(|_| anyhow::anyhow!("error log open worker failed"))??;
Ok(File::from_std(file))
}
fn format_error_event(event: &ErrorEvent, format: ErrorLogFormat) -> Vec<u8> {
match format {
ErrorLogFormat::Json => {
let mut object = Map::new();
object.insert("timestamp".into(), Value::String(event.timestamp.clone()));
object.insert("level".into(), Value::String("ERROR".into()));
object.insert("target".into(), Value::String(event.target.clone()));
object.insert("message".into(), Value::String(event.message.clone()));
for (key, value) in &event.fields {
if !matches!(key.as_str(), "timestamp" | "level" | "target" | "message") {
object.insert(key.clone(), Value::String(value.clone()));
}
}
let mut line = serde_json::to_vec(&Value::Object(object)).unwrap_or_default();
line.push(b'\n');
line
}
ErrorLogFormat::Raw => {
let one_line = |value: &str| value.replace('\r', "\\r").replace('\n', "\\n");
let mut line = format!(
"{} ERROR {} {}",
event.timestamp,
event.target,
one_line(&event.message)
);
for (key, value) in &event.fields {
line.push(' ');
line.push_str(key);
line.push('=');
line.push_str(&one_line(value));
}
line.push('\n');
line.into_bytes()
}
}
}
async fn error_log_loop(
mut receiver: mpsc::Receiver<ErrorEvent>,
settings: Arc<ArcSwap<ErrorLogSettings>>,
mut settings_changed: watch::Receiver<()>,
written: Arc<AtomicU64>,
errors: Arc<AtomicU64>,
) {
let mut open_path: Option<PathBuf> = None;
let mut file: Option<File> = None;
let mut retry_after: Option<Instant> = None;
let mut last_warning: Option<Instant> = None;
loop {
let event = tokio::select! {
event = receiver.recv() => match event {
Some(event) => event,
None => break,
},
changed = settings_changed.changed() => {
if changed.is_err() && receiver.is_closed() {
break;
}
let config = settings.load_full();
let path = PathBuf::from(&config.path);
if !config.enabled || open_path.as_ref() != Some(&path) {
// Dropping Tokio's File closes the descriptor without blocking
// the request path. A changed path is reopened on the next event.
file = None;
open_path = None;
retry_after = None;
}
continue;
}
};
let config = settings.load_full();
if !config.enabled {
file = None;
open_path = None;
retry_after = None;
continue;
}
let path = PathBuf::from(&config.path);
if open_path.as_ref() != Some(&path) {
file = None;
open_path = Some(path.clone());
retry_after = None;
}
if file.is_none() && retry_after.is_none_or(|until| Instant::now() >= until) {
match open_error_file(path.clone()).await {
Ok(opened) => {
file = Some(opened);
retry_after = None;
}
Err(error) => {
errors.fetch_add(1, Ordering::Relaxed);
retry_after = Some(Instant::now() + ERROR_LOG_RETRY);
if last_warning.is_none_or(|time| time.elapsed() >= ERROR_LOG_RETRY) {
tracing::warn!(path=%path.display(), %error, "optional error log file is unavailable");
last_warning = Some(Instant::now());
}
}
}
}
let Some(opened) = file.as_mut() else { continue; };
let line = format_error_event(&event, config.format);
if let Err(error) = opened.write_all(&line).await {
errors.fetch_add(1, Ordering::Relaxed);
file = None;
retry_after = Some(Instant::now() + ERROR_LOG_RETRY);
if last_warning.is_none_or(|time| time.elapsed() >= ERROR_LOG_RETRY) {
tracing::warn!(path=%path.display(), %error, "optional error log write failed");
last_warning = Some(Instant::now());
}
} else {
written.fetch_add(1, Ordering::Relaxed);
}
}
}
#[derive(Debug, Clone, Serialize)]
pub struct LogEntry {
pub vhost: Option<String>,
@@ -476,7 +884,8 @@ async fn sink_loop(
entry = rx.recv() => match entry { Some(entry) => entry, None => break },
};
let config = settings.load_full();
let payload = format_entry(&entry, &config.format);
let mut payload = format_entry(&entry, &config.format);
redact_value(&mut payload);
let mut attempted = false;
let mut failed = false;
if config.file.enabled {
@@ -716,6 +1125,38 @@ mod tests {
assert_eq!(value["response_bytes"], 42);
}
#[test]
fn error_log_defaults_off_and_redacts_common_secret_shapes() {
let settings = ErrorLogSettings::default();
assert!(!settings.enabled);
assert_eq!(settings.format, ErrorLogFormat::Json);
let text = redact_text(
r#"Authorization: Bearer topsecret url=https://x.invalid/?api_token=abc password=hunter2 {\"private_key\":\"pem-data\"}"#,
);
assert!(!text.contains("topsecret"));
assert!(!text.contains("api_token=abc"));
assert!(!text.contains("hunter2"));
assert!(!text.contains("pem-data"));
}
#[test]
fn error_json_is_one_record_and_reserved_fields_cannot_be_overwritten() {
let event = ErrorEvent {
timestamp: "2026-10-03T12:34:56.123Z".into(),
target: "rust_proxy::http".into(),
message: "upstream failed\nsecond line".into(),
fields: BTreeMap::from([
("message".into(), "attacker override".into()),
("request_id".into(), "abc".into()),
]),
};
let line = format_error_event(&event, ErrorLogFormat::Json);
assert_eq!(line.iter().filter(|byte| **byte == b'\n').count(), 1);
let value: Value = serde_json::from_slice(&line).unwrap();
assert_eq!(value["message"], "upstream failed\nsecond line");
assert_eq!(value["request_id"], "abc");
}
#[test]
fn retained_history_is_larger_than_the_live_snapshot_but_still_bounded() {
let logs = Logs::default();