The gateway already refused to ban addresses from proxy_protocol.trusted_cidrs,
but the traffic-intelligence auto-ban and the subnet monitor still did, so a
trusted load balancer living inside a banned /24 (or below the reputation
threshold) could be knocked out together with its clients.
- TrustedProxyCidr moved to src/config/trusted.rs (the edge section file hit
the 250-line gate) and gained `overlaps_prefix`: a ban is suppressed when the
banned prefix intersects a trusted network.
- TrafficHook::with_trusted suppresses reputation and offender bans; its ban()
now reports whether the ban was installed so `ban_offenders` counts real
bans.
- The subnet monitor takes a MonitorSetup bundle (spawn had 8 arguments) and
suppresses both userspace CIDR bans and XDP prefix bans that overlap a
trusted network, logging a warning instead.
- IPv4-mapped addresses are normalised in `contains` too.
- Tests: overlap maths (v4/v6, /0 rejected), tracker suppression vs unrelated
networks, and auto-ban skipping a trusted range while still banning public
offenders.
The subnet detector's spoof gate needs the number of distinct sources in a
prefix, but the XDP path reported 0, so a /24 flooded from a botnet could never
reach the Block verdict and the whole subnet detector was inert with XDP on.
- prefix_stats carries a 64-bit bitmap of hashed source addresses (interface
ID for IPv6, address for IPv4); userspace counts the set bits.
- read_prefix_stats now resets each entry after reading: the counters are
documented as per-window and compared against fixed thresholds, but were
cumulative, so with enough traffic every active prefix eventually looked
like an attack.
- The monitor no longer suppresses Block on the XDP path and no longer carries
the now-dead "no source diversity" branch.
ABI: the prefix_stats value grows 24 -> 32 bytes; the xdp_object_maps test
pins the new size.
Security:
- An empty API_PASSWORD passed validation (only the literal "changeme" was
rejected) and verify_password("", "") returned true, so a login with an
empty password minted an admin token. Require >= 16 characters and >= 32
non-whitespace bytes for JWT_SECRET; verify rejects an empty stored secret.
- syn_challenge_enabled required a real challenge_secret_hex: the public
placeholder from config.h is now rejected in every spelling.
- docker-compose: Redis/ClickHouse/Grafana are loopback-only and every
password uses ${VAR:?} instead of a default. An exposed Redis was a way to
lift bans and publish unbans to the whole fleet with one redis-cli.
- Empty CORS_ORIGIN no longer means "allow any origin".
- ban_cidr / userspace blacklist validate and mask prefixes; CIDR bans are
stored as ranges, so a /24 or /64 ban actually covers the subnet (before,
only the network address was blocked).
Availability:
- ClickHouse client has connect/request timeouts and no mutex held across
await: a hung analytics backend can no longer freeze the ban path. Events
go through a bounded buffer with a drop counter.
- Redis uses a ConnectionManager with per-command timeouts; pubsub and
reconcile reset the backoff after success; publish cannot block shutdown.
- Ban durations from pubsub are clamped and use checked_add; the sync task
survives panics instead of silently dying.
- Blacklist TOCTOU (remove_if) and replace_redis no longer empties the table
during a Redis reconnect.
- RateLimiter and reputation have hard caps with incremental eviction instead
of an unbounded DashMap and a stop-the-world retain.
- The attack detector scores before updating its baseline and needs several
quiet windows to leave UnderAttack, so a sustained flood keeps banning new
offenders instead of ending after one tick.
- accept() backs off on EMFILE/ENFILE, shutdown Err is treated as shutdown,
a global semaphore bounds in-flight connections, rate limiting runs before
PoW, whitelisted IPs are never banned, and connect/handshake/relay have
timeouts with an idle budget.
- PoW uses the configured difficulty as a floor, the clamp is enforced, and
nonce framing survives split segments and coalesced handshake bytes.
- Subnet tracker memory is capped (prefixes and per-prefix sources) and the
monitor bans with the configured TTL.
CI and docs:
- CI runs the test suite with --no-default-features --features protocol-http;
make ebpf fails on errors; the secrets gate catches weak compose defaults.
- doctor skips protected endpoints without a token and exits non-zero on
failure; a drain timeout returns exit code 1.
- TODO §2b records the review findings, what is fixed and what is deferred;
configuration/testing docs describe the new XDP and secret semantics.
XDP behaviour:
- udp_rate_limit6 (LRU, full 128-bit source key) plus throttle_hit6, so
udp_policy = "rate-limit" applies to native IPv6 as well.
- scripts/xdp_ipv6_smoke.sh: phase 4 patches .rodata before load and
verifies policy=2 on a real kernel (3 pass / 3 rate-limited).
Loader (XDP could never load before this):
- .rodata is one struct rampart_config with frozen offsets enforced by
_Static_assert; XdpGlobals is #[repr(C)] and the image is built with
offset_of!. Before, a 96-byte image was written into a 76-byte map at
offsets clang had reordered, so load() failed with EINVAL.
- patch_rodata fails loudly when the map size and image size diverge.
- build.rs tracks xdp/core and xdp/hooks, not just universal_filter.c.
- build_ringbuf looked for "events_map"; the map is "events_ringbuf".
- tests/xdp_object_maps.rs checks map names and .rodata size without root.
Docs and style:
- Code comments are English, JavaDoc/Rustdoc style, without emoji or
banners (new project rule, TODO §0.8).
- README rewritten bilingual: Russian first, English below, nav buttons,
compact TOC and copy-paste quick start.
- docs/README.md: documentation index and maintenance rules;
configuration.md documents the whole [xdp] section and IPv4/IPv6
coverage; testing.md covers the real-kernel smoke; stale links removed.