Include target_coords in proof_bytes() signed data to prevent transit
nodes from substituting fake coordinates. Add mandatory signature
verification at the originator using the target's public key from
identity_cache (guaranteed available since lookups are only initiated
from contexts where the key is already cached).
Verification failure discards the response. Identity cache miss (should
never happen) logs an error and discards. Add four new tests covering
verification success, failure, cache miss, and coordinate substitution
detection.
Rename FIPS Link Protocol (FLP) to FIPS Mesh Protocol (FMP)
The "Link Protocol" name understated the layer's scope — spanning tree
construction, bloom filter routing, greedy forwarding, and mesh-wide
coordination go well beyond link-level concerns. Rename fips-link-layer.md
to fips-mesh-layer.md, update FLP→FMP throughout docs and source code
(FLP_VERSION→FMP_VERSION, wire.rs, rx_loop.rs, spanning_tree.rs).
New SVG illustrations
- Protocol stack: color-coded layer diagram replacing ASCII art
- OSI mapping: side-by-side comparison with traditional networking layers
- Bloom filter propagation: 6-node tree with sender-colored filter boxes
showing split-horizon computation per link
- Routing decision flowchart: 5-step priority chain with candidate ranking
by tree distance and link performance
- Coordinate discovery: sequence diagram showing LookupRequest propagation,
response caching, and SessionSetup cache warming
Redesigned existing SVGs
- Architecture overview: uniform node layout, U-shaped encrypted link
connectors, separate end-to-end session line
- Node architecture: split Router Core into FSP and FMP layers, reorganize
transports into Overlay/Shared Medium/Point-to-Point categories
- Identity derivation: wider boxes, visible encode arrow, dashed npub line
fips-intro.md revisions
- Add inline references to prior work: Yggdrasil/Ironwood for coordinate
routing, Noise Protocol Framework for IK handshakes, WireGuard for
index-based session dispatch, Wikipedia for bloom filters, split-horizon,
and greedy embedding
- Add explanatory paragraphs after bloom filter diagram describing
split-horizon filter computation and candidate selection behavior
- Simplify transport abstraction language, remove I2P/LoRa references
- Fix LookupRequest wording ("propagates" not "floods"), note intermediate
node coordinate caching on lookup responses
- Rewrite architecture overview prose to match redesigned diagrams
SessionAck previously only carried the responder's coordinates
(src_coords). When the return path diverged from the forward path
(e.g., after tree reconvergence), transit nodes on the return path
lacked the initiator's coordinates and couldn't route the SessionAck
back, causing handshake timeouts.
Add dest_coords (initiator's coordinates) to the SessionAck wire
format, mirroring SessionSetup's design. Transit nodes now cache both
endpoints' coordinates when forwarding a SessionAck, making the return
path self-sufficient regardless of path asymmetry.
Root cause confirmed by churn-20 sim log analysis: the n04-n14
handshake failure was caused by n15 (return-path transit) lacking
n04's coordinates, not by stale tree routes through a downed node.
Auto-reconnect:
- Add per-peer auto_reconnect config (default true) to PeerConfig
- schedule_reconnect() feeds removed peers back into retry system with
unlimited retries and exponential backoff after MMP dead timeout
- RetryState gains reconnect flag to distinguish startup retries
(max_retries-limited) from auto-reconnect (unlimited)
Retry re-fire fix:
- process_pending_retries() now pushes retry_after_ms past the handshake
timeout window after successful initiate_peer_connection(), preventing
retries from firing every tick with no backoff
Chaos sim improvements:
- Directed outbound configs: BFS spanning tree + lower-ID-first assignment
eliminates dual-connect race conditions in simulation
- Save runner log (runner.log) alongside per-node logs for event correlation
- Increase churn-20 traffic aggressiveness and node churn (max_down_nodes
3→5, traffic interval min 0s, duration max 90s, concurrent flows 5→10)
handle_session_ack() had a bug where take_state() followed by
a non-Initiating match would re-insert the entry with state=None,
causing panics on subsequent .state()/.state_mut() calls. Fix by
checking is_initiating() before take_state().
Add safe is_initiating()/is_responding() to SessionEntry (matching
existing is_established()) and convert all production .state() callers
to use these None-safe methods. Gate .state() with #[cfg(test)].
Sends a 1-byte encrypted heartbeat (0x51) to each peer every 10s.
If no frame is received from a peer within 30s, the peer is removed
via remove_active_peer(), triggering tree reconvergence, coord cache
flush, and bloom filter recomputation.
This fixes the critical bug where UDP peers that silently died
(e.g., container stopped) were never detected or removed, leaving
the spanning tree permanently stale.
Both intervals are configurable via node.heartbeat_interval_secs
and node.link_dead_timeout_secs.
Add resend logic for SessionSetup/SessionAck messages routed through
the mesh. Stores the encoded payload on SessionEntry for resend in a
fresh SessionDatagram (so routing can adapt to topology changes).
Uses the same config parameters as link-layer retry.
Also fixes a latent bug: Initiating/Responding sessions previously
had no timeout — a stuck handshake would live forever. Now cleaned up
after handshake_timeout_secs (default 30s).
Responder idempotency: duplicate SessionSetup triggers resend of
stored SessionAck instead of being silently dropped. Initiator-side
duplicate SessionAck already handled safely (entry.take_state() sees
Established, puts it back and returns).
Handshake payload cleared on Established transition at both initiator
(handle_session_ack) and responder (handle_encrypted_session_msg).
Add message-level retry for Noise IK handshake within the 30s timeout
window. Previously, a lost msg1 or msg2 required the full timeout to
expire before cleanup and retry. Under 10% bidirectional loss (~19%
per attempt), this made connection establishment unreliable.
Initiator resends stored msg1 bytes with exponential backoff (1s, 2s,
4s, 8s, 16s — 5 resends). Responder stores msg2 and resends on
duplicate msg1 receipt. Duplicate msg2 at initiator drops silently
via existing pending_outbound cleanup.
Config: handshake_resend_interval_ms (1000), handshake_resend_backoff
(2.0), handshake_max_resends (5) on node.rate_limit.
P(all 6 attempts fail) under 19% loss = 0.19^6 ≈ 0.005%.
Log a warning when the kernel clamps SO_RCVBUF or SO_SNDBUF below
the requested size, directing the operator to increase rmem_max or
wmem_max. Document buffer size defaults and sysctl requirements in
the Docker config template.
Implement hybrid coordinate cache warming strategy: piggyback coords
via CP flag when they fit within transport MTU, send standalone
CoordsWarmup (0x14) message when they don't. On CoordsRequired or
PathBroken receipt, send CoordsWarmup immediately with source-side
rate limiting (default 2s per destination, configurable).
- Add CoordsWarmup = 0x14 session message type (empty body, CP flag)
- Add send_coords_warmup() following send_session_msg() pattern
- Restructure send_session_data() to send standalone warmup before
data packet when piggybacked coords exceed MTU
- Add immediate CoordsWarmup response in handle_coords_required()
and handle_path_broken() with per-destination rate limiting
- Add coords_response_interval_ms config (node.session)
- Add RoutingErrorRateLimiter::with_interval() constructor
- Zero transit-path changes: existing try_warm_coord_cache() handles
CoordsWarmup identically to CP-flagged data packets
- Update design docs (session layer, wire formats, mesh operation,
configuration)
FIPS_OVERHEAD was 150 but had two bugs: the session AEAD tag (16 bytes)
was listed in the comment but missing from the arithmetic, and the
coordinate budget (~60 bytes) was undersized and didn't belong in a
constant representing fixed data path overhead.
Corrected to 106 bytes (the actual fixed overhead without coordinates).
This increases effective_ipv6_mtu from 1322 to 1366 for standard
Ethernet, well above the IPv6 minimum of 1280.
Added a guard in send_session_data() that computes the total wire size
with coordinates before committing to include them. If adding coords
would exceed the transport MTU, the CP flag is skipped and the warmup
counter is not decremented. This prevents silently producing oversized
packets at tree depth 2+.
Updated design docs (ipv6-adapter, wire-formats, mesh-operation) with
corrected overhead values.
Under high-throughput forwarding, the kernel default 212KB receive
buffer overflows, silently dropping ~11% of UDP datagrams at transit
nodes. Add configurable recv_buf_size and send_buf_size to UdpConfig
(default 2MB each) using socket2 for pre-bind buffer configuration.
Startup log now reports actual buffer sizes granted by the kernel.
Requires net.core.rmem_max >= 2097152 on the host for the full 2MB
to take effect; otherwise the kernel silently clamps.
Periodic metrics: use long-term EWMA trends for RTT and loss instead
of instantaneous values, add jitter in ms, remove debug-level detail
block. Teardown metrics: add jitter and goodput, use consistent units
(ms for time, % for loss). Session periodic log shows observed MTU as
single 'mtu' field. RTT trend values correctly converted from
microseconds to milliseconds.
Replace raw NodeAddr hex strings in log output with human-readable
identifiers using a four-tier lookup: configured alias, active peer
short npub, session endpoint short npub, or truncated hex fallback.
- Add PeerIdentity::short_npub() for compact npub display (npub1xxxx...yyyy)
- Add NodeAddr::short_hex() for compact hex fallback (first 4 bytes + ...)
- Add peer_aliases map populated at startup from peer config
- Add Node::peer_display_name() with four-tier resolution
- Add SessionEntry::remote_pubkey() accessor for session-layer lookups
- Update all 120 log field occurrences across 11 handler/node files
- Pass pre-computed display names to MMP static metric/teardown methods
to work around borrow checker constraints in iterator loops
Per-packet happy-path events (UDP send/receive, TUN I/O, MMP report
processing, TreeAnnounce/FilterAnnounce sent, RTT samples) moved from
debug to trace. Periodic maintenance and retry scheduling moved from
info to debug. Session state changes (established, initiated, torn
down) and transport stop promoted from debug to info. MMP report send
failures demoted from warn to debug (normal under churn).
MMP reports (SenderReport, ReceiverReport, PathMtuNotification) were
resetting the idle timer on both RX and TX paths, preventing sessions
from ever timing out when MMP traffic kept flowing. Changed touch() to
only be called for DataPacket send/receive and session establishment,
so sessions with no application data tear down after the idle timeout
even with active MMP measurement traffic.
Session-layer encryption used an implicit nonce counter with no counter
on the wire, requiring packets to arrive in exact order. Under bulk
transfer load (e.g., SCP), UDP packet loss or reordering permanently
desynchronized sender/receiver counters, causing all subsequent
decryption to fail with no recovery.
Add an 8-byte counter field to the DataPacket wire format and switch
from implicit-counter decrypt() to decrypt_with_replay_check() which
uses the explicit wire counter plus a 2048-packet sliding replay
window — the same pattern already used at the link layer.
Wire format: msg_type(1) + flags(1) + counter(8) + payload_len(2) = 12
bytes (was 4). FIPS_OVERHEAD updated 127 → 135 bytes accordingly.
- PathBroken handler: convert to async, trigger re-discovery via
maybe_initiate_lookup(), reset COORDS_PRESENT warmup counter
(was a stub that only invalidated coord_cache)
- CoordsRequired recovery timing: reset warmup counter in
handle_lookup_response() when discovery completes for an
established session, so COORDS_PRESENT packets fire after
fresh coords are available (not just on CoordsRequired receipt)
- Routing error rate limiting: add RoutingErrorRateLimiter
(100ms per-destination, matching ICMP PTB pattern) to gate
send_routing_error() at transit nodes
- Remove root refresh dead code: the 1800s periodic root
re-announcement in check_tree_state() only propagated to
depth 1 (sequence-only changes don't cascade). Root loss
detection relies on link failure propagation which works
correctly.
Include source and destination coordinates on the first N DataPackets
of each session (default 5, configurable via
node.session.coords_warmup_packets). This warms transit node
coord_caches so multi-hop forwarding and error signal routing work
after SessionSetup cache entries expire.
On CoordsRequired receipt, reset the counter to re-enable coordinate
inclusion for the next N packets, handling mid-session cache expiry
and path changes.
Changes:
- SessionConfig: add coords_warmup_packets field (default 5)
- SessionEntry: add coords_warmup_remaining counter, initialized on
Established transition (both initiator and responder paths)
- send_session_data(): attach coords via DataPacket::with_coords()
while counter > 0, decrement per send
- handle_coords_required(): reset counter for affected session
- 4 new unit tests for counter lifecycle and config default
Identity cache: remove TTL-based expiry (60s TTL broke active sessions
after expiry since handle_tun_outbound checks identity_cache before
session table). Replace with LRU-only eviction bounded by configurable
identity_size (default 10K). Lookup now touches timestamp for LRU
freshness.
Cache merge: unify coord_cache and route_cache into single coordinate
cache. Both stored NodeAddr→TreeCoordinate; the layer distinction was
conceptual, not functional. Discovery-sourced entries now get the same
TTL+refresh treatment as session-sourced entries. Simplifies
find_next_hop() to single cache lookup.
Parent-change flush: clear coord_cache after recompute_coords() in both
parent-switch paths of handle_tree_announce(). Stale coordinates after
tree reconvergence cause dead-end routing that's more expensive than
re-discovery.
Tested: 493 unit tests passed, clippy clean, Docker mesh 20/20,
Docker chain 6/6.
Add dual-approach MTU handling to prevent TCP connections from hanging
when packets exceed the transport MTU after FIPS encapsulation.
ICMPv6 Packet Too Big:
- Generate RFC 4443 PTB messages for oversized packets at TUN outbound
- Inject back via TUN for local delivery to the application
- Per-source rate limiting (100ms interval, 10s entry expiry)
- MTU check in handle_tun_outbound before session encapsulation
TCP MSS Clamping:
- Intercept SYN packets in run_tun_reader() (outbound)
- Intercept SYN-ACK packets in TunWriter (inbound)
- Clamp MSS option to fit within effective MTU (transport - 127 overhead)
- Recalculate TCP checksum after modification
Code organization:
- ICMP, TCP MSS, and rate limiter modules in upper/ alongside existing
protocol-specific packet handling (dns.rs, tun.rs)
- Shared FIPS_OVERHEAD constant (127 bytes) and effective_ipv6_mtu()
function in upper/icmp.rs
- Node::effective_ipv6_mtu() delegates to the shared function
- run_tun_reader() accepts actual transport MTU from config
Example config corrections:
- UDP transport MTU set to 1472 across all configs (correct max UDP
payload for standard Ethernet: 1500 - 20 IPv4 - 8 UDP)
- Startup logging of effective MTU and max MSS values
Sessions in the Established state that have no activity for 90 seconds
are now automatically removed. This ensures idle sessions are torn down
before transit node coord_cache entries expire (300s TTL), so that when
traffic resumes a fresh SessionSetup re-warms transit node caches with
current coordinates.
The identity cache now stores registration timestamps and expires
entries after 60 seconds via lazy expiry on lookup. This prevents
unbounded growth while allowing natural repopulation through DNS
resolution on next use.
Timer ordering: identity (60s) < session (90s) < coord_cache (300s).
Both timeouts are configurable: node.session.idle_timeout_secs and
node.cache.identity_ttl_secs. Setting idle_timeout_secs to 0 disables
session idle purging.
Changes:
- Add idle_timeout_secs (default 90) to SessionConfig
- Add identity_ttl_secs (default 60) to CacheConfig
- Add timestamp to identity_cache entries, lazy expiry on lookup
- Add purge_idle_sessions() called from tick loop
- Remove #[cfg(test)] from SessionEntry::last_activity()
- 7 new tests covering timeout behavior and edge cases
Transit nodes cache destination coordinates when they forward
SessionSetup messages (via try_warm_coord_cache). These coord_cache
entries have a 5-minute TTL, after which they expire. Once expired,
the transit node can no longer forward data packets for that
destination — find_next_hop returns None and the node sends
CoordsRequired errors back to the source. This creates a permanent
routing failure for any multi-hop path after 5 minutes of the initial
session establishment, even if traffic is actively flowing.
The root cause was that find_next_hop used coord_cache.get(), a
read-only lookup that checks expiry but never extends it. Active
forwarding did not keep the cache warm. Meanwhile, get_and_touch()
existed but only updated last_used without extending expires_at.
Fix:
- find_next_hop now calls coord_cache.get_and_touch() instead of get()
- get_and_touch now calls entry.refresh() instead of entry.touch(),
which extends expires_at by the default TTL on each access
- find_next_hop signature changed from &self to &mut self to allow
the mutable cache access
This ensures that as long as traffic flows through a transit node,
the coord_cache entries stay warm and routing continues to work.
Entries still expire after 5 minutes of inactivity as designed.
FilterAnnounce messages never settled in steady state because
handle_filter_announce unconditionally marked all peers for
re-announcement on every inbound filter, creating a perpetual
ping-pong at ~1 message/sec.
Added last_sent_filters tracking to BloomState. mark_changed_peers()
computes outgoing filters and compares against what was last sent,
only marking peers whose filter content actually differs.
Wire the discovery protocol into the data plane path so that
find_next_hop() consults route_cache as a fallback when coord_cache
has no entry. When session initiation fails due to missing routes,
trigger discovery (initiate_lookup) instead of immediately sending
ICMPv6 Destination Unreachable. On discovery completion, retry
session initiation for any pending TUN packets.
Changes:
- find_next_hop() falls back to route_cache when coord_cache misses
- handle_tun_outbound() triggers discovery on session failure
- handle_lookup_response() retries session after discovery completes
- handle_coords_required() triggers discovery for missing coordinates
- Add pending_lookups deduplication with 10-second timeout
- Periodic cleanup of stale lookups in tick handler
- New test: route_cache fallback verification in find_next_hop
Add fd00::/8 route to TUN configure_interface() via rtnetlink - without
this route the kernel had no path to FIPS addresses ("network unreachable").
Update two-node-udp README based on live namespace testing:
- Remove resolvectl steps (doesn't work inside namespaces)
- Remove ICMPv6 Echo Reply mention (kernel handles it natively)
- Add namespace DNS limitation note and kernel ping reply explanation
- Simplify from 8 to 7 steps
Add DNS responder that resolves <npub>.fips queries to FipsAddress IPv6
addresses. Resolution is pure computation (npub → NodeAddr → IPv6) with
identity cache priming as a side effect, enabling subsequent TUN packet
routing to non-peer destinations.
- New dns.rs module: resolve_fips_query(), handle_dns_packet() using
simple-dns crate, run_dns_responder() async UDP server loop
- DnsConfig in config.rs: enabled, bind_addr (127.0.0.1), port (5354)
- Fourth select! arm in RX event loop for DNS identity channel
- DNS task spawn/abort in node lifecycle
- 10 new tests (420 total)
Add examples/two-node-udp/ with standalone walkthrough for testing two
FIPS nodes in Linux network namespaces over a veth pair. Includes
network diagram, config files, DNS routing setup, and troubleshooting.
Wire the TUN reader to the session layer for end-to-end IPv6 packet
delivery through the mesh. TUN reader forwards FIPS-destined packets
(fd::/8) to Node via tokio::sync::mpsc channel. Identity cache maps
FipsAddress prefix bytes to NodeAddr + PublicKey for reverse address
resolution, populated at peer promotion and session creation. Outbound
handler validates IPv6, looks up destination identity, routes through
established session or initiates new one with pending packet queue
(16/dest, 256 dests max). Queue flushed on session establishment.
ICMPv6 Destination Unreachable for unknown destinations. 6 new tests
including 3-node forwarded TUN data and pending queue flush. 410 tests
pass, clean build.
Implement Noise IK session handshake between arbitrary endpoints, carried
inside SessionDatagram envelopes through the mesh. Sessions use a three-state
machine (Initiating → Established for initiator, Responding → Established
for responder on first DataPacket). Includes session initiation API,
encrypted data transfer, simultaneous initiation tie-break, error signal
handlers (CoordsRequired, PathBroken), and local delivery wiring in the
forwarding handler.
New files: node/session.rs (state types), handlers/session.rs (~500 lines,
all session message handlers + send path), tests/session.rs (11 tests).
100-node integration test establishes sessions across random topology,
sends bidirectional encrypted datagrams through injected TUN channels,
and verifies 200/200 deliveries with 100% session establishment. Reports
routing statistics including avg 4.1 link hops per datagram.
404 tests pass (up from 393).
Add handle_session_datagram handler replacing the 0x40 dispatch stub.
Transit nodes now decode the datagram envelope, enforce hop limits,
warm coordinate caches from SessionSetup/SessionAck/DataPacket payloads,
route via find_next_hop, and generate CoordsRequired/PathBroken error
signals on routing failure.
14 new tests covering decode errors, hop limit enforcement, local
delivery, cache warming for all session message types, single-hop and
multi-hop forwarding through live node chains, error signal generation,
and cache warming enabling subsequent routing. 375 tests pass.
Add src_addr to SessionDatagram envelope (34-byte header: msg_type +
src_addr + dest_addr + hop_limit) so transit routers can route error
signals back to the packet's originator.
Reclassify CoordsRequired/PathBroken as link-layer error signals
(plaintext inside SessionDatagram) rather than e2e encrypted session
messages. Transit routers generate these when forwarding fails and
route them to src_addr; if source is also unreachable, drop silently.
Remove redundant src_addr/dest_addr/hop_limit from SessionSetup,
SessionAck, and DataPacket (now in envelope). DataPacket header
reduced from 36 to 4 bytes. Remove PathBroken.original_src.
Fix routing loop vulnerability: gate bloom filter path on having
cached dest_coords to prevent blind forwarding between peers.
Simplify select_best_candidate() to require coordinates.
Fix gossip protocol type codes (0x11->0x20, 0x12->0x30, 0x13->0x31)
for consistency across all design docs.
All 5 design docs updated and cross-checked for consistency.
335 tests pass, zero warnings.
Add the full next-hop routing algorithm to Node::find_next_hop():
- Local delivery, direct peer, bloom filter candidates, greedy tree
routing fallback, with (link_cost, tree_distance, node_addr) ordering
- select_best_candidate() scores by peer→dest distance (not us→peer)
with self-distance check to prevent routing loops
- TreeState::find_next_hop() for greedy tree routing with progress
guarantee
- ActivePeer::link_cost() placeholder (constant 1.0) for future link
quality metrics
Add routing tests including 100-node all-pairs reachability simulation
(9900/9900 delivered, 0 loops, avg 4.0 hops, max 8).
Update fips-routing.md to reflect bloom filter routing as the primary
forwarding mechanism, with greedy tree routing as fallback during
convergence windows.
Split handlers.rs (986 lines) into handlers/ with 5 subfiles organized
by responsibility: rx_loop, encrypted, handshake, dispatch, timeout.
Split tests.rs (2350 lines) into tests/ with 4 subfiles: unit tests,
handshake integration, spanning tree convergence, and bloom filter tests.
Shared test helpers extracted to tests/mod.rs.
Visibility adjusted from pub(super) to pub(in crate::node) for handler
methods now two levels deep. Unused imports cleaned up in node/mod.rs.
All 316 tests pass, zero warnings.
When both nodes simultaneously initiate connections, each promotes its
inbound handshake first, leaving both sides with mismatched Noise sessions
and session indices. Resolve this in handle_msg2 using the existing
tie-breaker (smaller node_addr's outbound wins):
- Winner (smaller node): swaps ActivePeer to outbound session + indices
via new replace_session() method
- Loser (larger node): keeps inbound session and original their_index
(already references winner's outbound our_index from msg1)
Key insight: the loser must NOT update their_index from msg2, because the
msg2 sender_idx carries the winner's inbound index which becomes stale
after the winner swaps.
Defer outbound connection cleanup from promote_connection() to handle_msg2()
so outbound session state is available for the swap.
Verified with live two-node test over UDP in network namespaces: both nodes
connect, exchange TreeAnnounce, and converge to spanning tree within ~2s.
Multi-node integration test infrastructure with 100-node random topology
convergence test (5 topology variants). Two-phase drain loop processes
Noise IK handshakes and TreeAnnounce cascades over live UDP until
quiescence.
Fix parent ancestry propagation in handle_tree_announce: when a node's
current parent updates its tree state (e.g., learning a new root from
upstream), coordinates must be recomputed even though the parent itself
didn't change. Without this, chains longer than 2 hops fail to converge.
Johnathan Corgan
Origami74andJohnathan Corgan