This commit introduces a comprehensive set of improvements to enhance
the security, reliability, and configurability of the proxy server,
specifically targeting adversarial resilience and high-load concurrency.
Security & Cryptography:
- Zeroize MTProto cryptographic key material (`dec_key`, `enc_key`)
immediately after use to prevent memory leakage on early returns.
- Move TLS handshake replay tracking after full policy/ALPN validation
to prevent cache poisoning by unauthenticated probes.
- Add `proxy_protocol_trusted_cidrs` configuration to restrict PROXY
protocol headers to trusted networks, rejecting spoofed IPs.
Adversarial Resilience & DoS Mitigation:
- Implement "Tiny Frame Debt" tracking in the middle-relay to prevent
CPU exhaustion from malicious 0-byte or 1-byte frame floods.
- Add `mask_relay_max_bytes` to strictly bound unauthenticated fallback
connections, preventing the proxy from being abused as an open relay.
- Add a 5ms prefetch window (`mask_classifier_prefetch_timeout_ms`) to
correctly assemble and classify fragmented HTTP/1.1 and HTTP/2 probes
(e.g., `PRI * HTTP/2.0`) before routing them to masking heuristics.
- Prevent recursive masking loops (FD exhaustion) by verifying the mask
target is not the proxy's own listener via local interface enumeration.
Concurrency & Reliability:
- Eliminate executor waker storms during quota lock contention by replacing
the spin-waker task with inline `Sleep` and exponential backoff.
- Roll back user quota reservations (`rollback_me2c_quota_reservation`)
if a network write fails, preventing Head-of-Line (HoL) blocking from
permanently burning data quotas.
- Recover gracefully from idle-registry `Mutex` poisoning instead of
panicking, ensuring isolated thread failures do not break the proxy.
- Fix `auth_probe_scan_start_offset` modulo logic to ensure bounds safety.
Testing:
- Add extensive adversarial, timing, fuzzing, and invariant test suites
for both the client and handshake modules.
- Bump versions of several dependencies in Cargo.toml for improved functionality and security, including:
- socket2 to 0.6
- nix to 0.31
- toml to 1.0
- x509-parser to 0.18
- dashmap to 6.1
- rand to 0.10
- reqwest to 0.13
- notify to 8.2
- ipnetwork to 0.21
- webpki-roots to 1.0
- criterion to 0.8
- Introduce `OnceLock` for secure random number generation in multiple modules to ensure thread safety and reduce overhead.
- Refactor random number generation calls to use the new `RngExt` trait methods for consistency and clarity.
- Add new PNG files for architectural documentation.
- Introduced red-team expected-fail tests for client masking shape hardening.
- Added integration tests for masking AB envelope blur to improve obfuscation.
- Implemented masking security tests to validate the behavior of masking under various conditions.
- Created tests for masking shape above-cap blur to ensure proper functionality.
- Developed adversarial tests for masking shape hardening to evaluate robustness against attacks.
- Added timing normalization security tests to assess the effectiveness of timing obfuscation.
- Implemented red-team expected-fail tests for timing side-channel vulnerabilities.
- Introduced multiple adversarial tests for MTProto handshake to ensure robustness against replay attacks, invalid mutations, and concurrent flooding.
- Implemented a function to build proxy headers based on the specified version, improving the handling of masking protocols.
- Added tests to validate the behavior of the masking functionality under various conditions, including unknown proxy protocol versions and oversized payloads.
- Enhanced relay tests to ensure stability and performance under high load and half-close scenarios.
- Introduced a new test module for middle relay idle policy security tests, covering various scenarios including soft mark, hard close, and grace periods.
- Implemented functions to create crypto readers and encrypt data for testing.
- Enhanced the Stats struct to include counters for relay idle soft marks, hard closes, pressure evictions, and protocol desync closes.
- Added corresponding increment and retrieval methods for the new stats fields.
- Per-user ad_tag in [access.user_ad_tags], global fallback in general.ad_tag
- User tag overrides global; if no user tag, general.ad_tag is used
- Both general.ad_tag and user_ad_tags support hot-reload (no restart)
Alexey