mrambossekandClaude Fable 5
a7dccf7da2
Letting the kernel fragment an oversized datagram answers one question — do
fragments get through. It cannot answer the more interesting one, because the
kernel always emits them in order, first one first.
The classic middlebox fault is exactly about that ordering. Only the first
fragment carries the UDP header, and therefore the ports; a stateful firewall
or NAT that has not seen it has no flow to match the rest against, and many
drop them. That is invisible to any in-order test and shows up in the field as
"large DNS answers fail on this network" or "the tunnel breaks when the MTU
drops" — it works until the network reorders, then fails intermittently, which
is the hardest kind of fault to chase.
So the server now builds the fragments itself (raw socket, IP_HDRINCL) and
controls their order: in_order as a baseline, reversed, and first-fragment-last.
The datagram is assembled and signed whole before being cut up, so what the
client reassembles is indistinguishable from an ordinary packet — otherwise it
would be measuring our sender rather than the path.
Two details that would silently produce wrong answers:
- The UDP checksum is computed rather than left zero. A zero-checksum datagram
is dropped by some middleboxes, and that drop would be recorded as a
fragmentation failure, which is the wrong conclusion entirely.
- Fragment offsets are in 8-byte units, so non-final fragments are rounded to
a multiple of 8. A 100-byte fragment is not an error, it is a datagram no
host will ever reassemble.
frag-send is advertised only when a raw socket can actually be opened — checked
by opening one, since a permission model has more ways to say no than a
capability bit has to say yes.
Fragment header arithmetic is unit-tested (reassembly coverage, MF flags, shared
IP ID, 8-byte offsets, checksum verification), cross-compiled and run on Linux
since the code is build-tagged.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Echolot app
The production Android client (spec). Native Kotlin + Jetpack Compose. Multi-module; built bottom-up from a verifiable protocol spine.
Modules
| Module | Type | Status |
|---|---|---|
core-protocol |
pure Kotlin/JVM | done — client half of probe-protocol.md, verified live against the server |
core-measurement |
pure Kotlin/JVM | planned — measurement-schema.md types |
core-probe |
Android lib | planned — app-tier probes, ported from echolot-prober |
core-shizuku |
Android lib | planned — dual-path executor (UserService + newProcess fallback) |
app |
Android app | planned — Compose UI |
core-protocol is deliberately Android-free so it builds and unit-tests on any JDK (no Android
SDK) and can run integration tests against a live server.
core-protocol
Implements the control plane (SPKI-pinned enrollment/profile/sessions via HttpsURLConnection —
Android-API-1 compatible, hostname verification off because trust is the pin), the HKDF-SHA256
session-key schedule, and the binary ELT1 UDP data plane (HMAC gate, ECHO + observation block,
MTU probe) — byte-compatible with the Go server.
./gradlew :core-protocol:test # unit tests (crypto vectors, wire round-trip)
scripts/test-fmr.sh # live end-to-end test against the deployed server
test-fmr.sh mints an enrollment token over SSH, enrolls via the public control plane, computes
the SPKI pin from the served cert, and runs LiveServerTest — proving the client speaks the wire
protocol to the real server (enroll → profile → session → echo+observation → MTU → observations).
The live test self-skips when ECHOLOT_LIVE_* env vars are absent, so unit runs and CI stay green
offline.