Johnathan Corgan ec64a0dce1 Add TCP transport implementation and test harness support
Implement TCP transport for FIPS enabling firewall traversal and serving
as the foundation for future Tor transport. This is the first
connection-oriented transport in the system.

Key design decisions:
- FMP header-based framing: reuses existing 4-byte FMP common prefix for
  packet boundary recovery with zero framing overhead
- Session survives TCP reconnection: Noise/MMP/FSP state bound to npub,
  not TCP connection; MMP liveness is sole authority for peer death
- Connect-on-send: fresh connection on first send, transparent reconnect
- close_connection() trait method for cross-connection deduplication cleanup

New transport files:
- src/transport/tcp/mod.rs: TcpTransport, connection pool, accept loop
- src/transport/tcp/stream.rs: FMP-aware stream reader (shared with Tor)

Modified: transport trait (close_connection), TcpConfig, TransportHandle
match arms, create_transports(), initiate_connection() for connection-
oriented links, cross-connection tie-breaker cleanup, design docs.

Tree announce loop and TCP stability fixes:
- Preserve tree announce rate-limit state across reconnection: carry
  forward last_tree_announce_sent_ms when a peer reconnects so the
  rate-limit window isn't reset to zero
- Drop oversize TCP packets at sender: pre-send MTU check returns
  MtuExceeded instead of writing to the stream, preventing receiver-side
  connection teardown and reset-reconnect cycles

Chaos harness:
- TCP transport support: tcp_edges/has_tcp/tcp_peers in SimTopology,
  transport-aware config_gen with per-edge transport type, TCP port 443,
  pure-TCP node support
- Include all non-Ethernet edges in directed_outbound()
- Fix netem/links log messages to say "IP-based" instead of "UDP"
- Add tcp-chain, tcp-only, and tcp-mesh scenario files

Static harness:
- Transport-aware config generation (get_default_transport, transport_port)
- TCP transport injection via Python post-processing
- Add tcp-chain topology and docker-compose profile
2026-02-27 00:41:37 +00:00
2026-02-22 20:52:55 +00:00

FIPS: Free Internetworking Peering System

License: MIT Rust Status

A distributed, decentralized network routing protocol for mesh nodes connecting over arbitrary transports.

Status: Alpha (0.1.0)

FIPS is under active development. The protocol and APIs are not stable. Expect breaking changes. See Status & Roadmap below.

Overview

FIPS is a self-organizing mesh network that operates natively over a variety of physical and logical media — local area networks, Bluetooth, serial links, radio, or the existing internet as an overlay. Nodes generate their own identities, discover each other, and route traffic without any central authority or global topology knowledge.

FIPS uses Nostr keypairs (secp256k1/schnorr) as native node identities, making every Nostr user a potential network participant. Nodes address each other by npub, and the same cryptographic identity used in the Nostr ecosystem serves as both the routing address and the basis for end-to-end encrypted sessions across the mesh.

Features

  • Self-organizing mesh routing — spanning tree coordinates and bloom filter candidate selection, no global routing tables
  • Multi-transport — UDP/IP overlay today; designed for Ethernet, Bluetooth, serial, radio, and Tor
  • Noise encryption — hop-by-hop link encryption plus independent end-to-end session encryption
  • Nostr-native identity — secp256k1 keypairs as node addresses, no registration or central authority
  • IPv6 adaptation — TUN interface maps npubs to fd00::/8 addresses for unmodified IP applications
  • Metrics Measurement Protocol — per-link RTT, loss, jitter, and goodput measurement
  • Operator visibilityfipsctl control socket interface for runtime inspection of peers, links, sessions, tree state, and metrics
  • Zero configuration — sensible defaults; a node can start with no config file, though peer addresses are needed to join a network

Quick Start

Requirements

  • Rust 1.85+ (edition 2024)
  • Linux (TUN interface requires CAP_NET_ADMIN or root)

Build

git clone https://github.com/jmcorgan/fips.git
cd fips
cargo build --release

Run

# Start with default search paths (see below):
sudo ./target/release/fips

# With an explicit configuration file:
sudo ./target/release/fips -c fips.yaml

Without -c, the node searches for fips.yaml in these locations (highest priority first, values from later files override earlier ones):

  1. ./fips.yaml (current directory)
  2. ~/.config/fips/fips.yaml (user config)
  3. /etc/fips/fips.yaml (system)

If no config file is found, the node starts with defaults (ephemeral identity, default ports, no peers).

A minimal two-node setup (each node points at the other):

# node-a.yaml                          # node-b.yaml
node:                                  # node:
  identity:                            #   identity:
    nsec: "nsec1aaa..."                #     nsec: "nsec1bbb..."
transports:                            # transports:
  udp:                                 #   udp:
    bind_addr: "0.0.0.0:2121"          #     bind_addr: "0.0.0.0:2121"
peers:                                 # peers:
  - npub: "npub1bbb..."                #   - npub: "npub1aaa..."
    addresses:                         #     addresses:
      - transport: udp                 #       - transport: udp
        addr: "10.0.0.2:2121"          #         addr: "10.0.0.1:2121"

The nsec field accepts bech32 (nsec1...) or hex-encoded secret keys. Omit it entirely for an ephemeral identity that changes each restart.

See docs/design/fips-configuration.md for the full configuration reference.

Test Connectivity

FIPS includes a built-in DNS resolver (enabled by default, port 5354) that maps .fips names to fd00::/8 IPv6 addresses derived from each node's public key. Configure your system to send .fips queries to it.

With systemd-resolved:

sudo resolvectl dns fips0 127.0.0.1:5354
sudo resolvectl domain fips0 ~fips

Or manually in /etc/resolv.conf (routes all DNS through FIPS for .fips names only if your resolver supports conditional forwarding; otherwise this sets it as a general nameserver):

nameserver 127.0.0.1
options port:5354

Once DNS is configured, ping a peer by npub:

ping6 npub1bbb....fips

Any IPv6-capable application can reach FIPS nodes this way — ping6, ssh, curl, etc.

Inspect

While a node is running, use fipsctl to inspect its state:

fipsctl show status       # Node status overview
fipsctl show peers        # Authenticated peers
fipsctl show links        # Active links
fipsctl show tree         # Spanning tree state
fipsctl show sessions     # End-to-end sessions
fipsctl show bloom        # Bloom filter state
fipsctl show mmp          # MMP metrics summary
fipsctl show cache        # Coordinate cache stats
fipsctl show connections  # Pending handshake connections
fipsctl show transports   # Transport instances
fipsctl show routing      # Routing table summary

fipsctl communicates with the node via a Unix domain control socket (enabled by default). All queries are read-only. Use -s <path> to override the socket path.

Multi-node Testing

See testing/ for Docker-based integration test harnesses including static topology tests and stochastic chaos simulation.

Documentation

Protocol design documentation is in docs/design/, organized as a layered protocol specification. Start with fips-intro.md for the full protocol overview.

Project Structure

src/          Rust source (library + fips/fipsctl binaries)
docs/design/  Protocol design specifications
testing/      Docker-based integration test harnesses

Status & Roadmap

FIPS is at v0.1.0 (alpha). The core protocol works end-to-end over UDP/IP overlays but has not been tested beyond small meshes.

What works today

  • Spanning tree construction with greedy coordinate routing
  • Bloom filter discovery for finding nodes without global state
  • Noise IK (link layer) and Noise XK (session layer) encryption
  • IPv6 TUN adapter with DNS resolution of .fips names
  • Per-link metrics (RTT, loss, jitter, goodput)
  • Runtime inspection via fipsctl
  • Docker-based integration and chaos testing

Near-term priorities

  • Peer discovery via Nostr relays (bootstrap without static peer lists)
  • Additional transports (Ethernet, Tor)
  • Improved routing resilience under churn
  • Security audit of cryptographic protocols
  • CI pipeline and published crate

Longer-term

  • Mobile platform support
  • Bandwidth-aware routing and QoS
  • Protocol stability and versioned wire format

License

MIT — see LICENSE.

Description
The Free Internetworking Peering System
Readme MIT 30 MiB
Languages
Rust 86.1%
Shell 10.4%
Python 3%
PowerShell 0.2%
Makefile 0.1%