popov/neuron-tai
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: harden DGR-002 protocol bounds

Browse Source
This commit is contained in:
Dobromir Popov
2026-07-13 17:30:54 +03:00
parent 30dcf953fe
commit d904c40f66
10 changed files with 470 additions and 63 deletions
Download Patch File Download Diff File Expand all files Collapse all files

69
.scratch/distributed-gguf-runtime/evidence/DGR-002/README.md
View File

@@ -38,7 +38,7 @@ New:
| Path | What |
|---|---|
| `packages/node/native/proto/shard_runtime.proto` | The schema (sha256 `077ee349…`, see `protocol.json`) |
| `packages/node/native/proto/shard_runtime.proto` | The schema (sha256 `9e211660…`, see `protocol.json`) |
| `packages/node/native/CMakeLists.txt` | C++ generation + build wiring + ctest |
| `packages/node/native/tests/test_shard_protocol_conformance.cpp` | C++ conformance test |
| `packages/node/native/testdata/*.binpb` | Committed cross-language vectors |
@@ -50,14 +50,19 @@ New:
| `scripts/generate_native_protocol.py` | Python generation, with `--check` |
| `scripts/generate_protocol_goldens.py` | Vector generation, with `--check` |
| `scripts/bootstrap_native_toolchain.sh` | Builds protobuf C++ from source |
| `tests/test_native_shard_protocol.py` | 29 Python tests |
| `tests/test_native_shard_protocol.py` | 45 Python tests |
Modified:
- `packages/node/pyproject.toml` — added `grpcio>=1.60`, `protobuf>=5`; new `proto`
extra pinning `grpcio-tools==1.82.1`.
- `packages/node/pyproject.toml` — added runtime floors `grpcio>=1.82.1` and
`protobuf>=7.35.0`, matching the committed generated-code requirements; new
`proto` extra pinning `grpcio-tools==1.82.1`.
- `packages/node/meshnet_node/activation_compression.py` — optional bounded zstd
output for untrusted protocol frames; existing callers remain compatible.
- `packages/node/meshnet_node/native_protocol/__init__.py` — exports negotiated
bound constants and whole-session-message validation.
No other working-tree file was touched. `git status` before this story was clean.
The canonical PRD marks only DGR-002 passed. `git status` before this story was clean.
## 3. Commands and real results
@@ -76,14 +81,45 @@ ctest --test-dir build/native --output-on-failure -> 1/1 Test #1: shard_prot
cmp build/native/cpp_roundtrip.binpb \
packages/node/native/testdata/session_request_golden.binpb -> identical (exit 0)
pytest -q tests/test_native_shard_protocol.py -> 29 passed
pytest -q (full suite) -> 712 passed, 12 skipped
pytest -q tests/test_native_shard_protocol.py -> 45 passed
pytest -q tests/test_native_shard_protocol.py \
tests/test_activation_compression.py -> 51 passed
pytest -q (final full suite) -> 728 passed, 12 skipped
pytest -q tests/test_tracker_routing.py::test_tracker_dashboard_can_cancel_inflight_proxy
(after an earlier flaky full-suite failure) -> 1 passed, 1 passed, 1 passed
clean minimum-runtime import + codec smoke test -> passed
grpcio==1.82.1, protobuf==7.35.0
compileall -q packages tests -> OK (exit 0)
git diff --check -> clean (exit 0)
```
The C++ lane was rebuilt from scratch (`rm -rf build/native`) using only the
documented commands, and reproduced the same result.
The C++ lane was rebuilt from scratch by Ralph (`rm -rf build/native`) using only
the documented commands, and reproduced the same result. During controller
review the user explicitly chose not to repeat the destructive build-directory
cleanup, so the independent controller relied on the recorded CMake/CTest run
while reproducing every Python/generation/full-suite gate.
### Controller review corrections
Independent controller review found and fixed two classes of issue before
integration:
1. Generated stubs required gRPC 1.82.1 and Protobuf 7.35.0, while the initial
package metadata allowed much older runtimes that could fail at import time.
2. Flow-control bounds were described but not enforced by the reference decoder.
Tensor declarations, shape rank/dimensions, fragment/tensor counts, fragments,
wire bodies, whole bundles, complete session messages (including envelope
overhead), and zstd window/output expansion are now fail-closed against the
negotiated/default bounds. Unspecified bundle versions, compression and
checksums are rejected rather than interpreted as valid data.
3. Negotiated initial credits could exceed `max_inflight_chunks`; credits are now
capped by the settled in-flight limit.
Controller results: protocol tests `45 passed`; protocol plus shared compression
tests `51 passed`; final full suite `728 passed, 12 skipped`. A clean environment
at the declared minimum gRPC/Protobuf runtime versions imported both generated
stub modules and round-tripped the codec. Generation checks, `compileall`, static
secret scan, and `git diff --check` all passed.
### Full-suite note — a pre-existing flaky test
@@ -104,10 +140,11 @@ pytest -q tests/test_tracker_routing.py::test_tracker_dashboard_can_cancel_infli
It is a timing race in proxy cancellation (a 3-second in-flight generation raced
against the cancel assertion), not a deterministic failure, and it touches no code
this story changes. The final full-suite run with DGR-002 applied was green
(`712 passed, 12 skipped`) — 683 pre-existing tests plus this story's 29. Flagged
for whoever owns the tracker cancel path; **not** fixed here, since silently
touching another story's code is out of scope.
this story changes. One controller full-suite run reported exactly that one failure
(`1 failed, 719 passed, 12 skipped`); three immediate isolated retries all passed
in 1.11 seconds, and the final exact-code full suite was green (`728 passed,
12 skipped`). It is flagged for whoever owns the tracker cancel path and is **not**
fixed here, since silently touching another story's code is out of scope.
## 4. Acceptance criteria
@@ -118,12 +155,12 @@ touching another story's code is out of scope.
| Bounded chunking for prefill; small decode fast path | `ChunkInfo` + `plan_prefill_chunks` (128-token bound, ADR-0008); `DecodeStep`; `test_prefill_is_split_into_bounded_token_aligned_chunks`, `test_decode_fast_path_is_much_smaller_than_a_full_envelope_chunk` |
| Envelope carries schema version, work id, session id, epoch, fingerprint, range/effective start, phase, position, idempotency step, cache expectation, compression, checksum | `Envelope` + `NamedTensor`; `test_envelope_carries_every_field_the_protocol_promises` asserts against the **descriptor**, so deleting a field from the `.proto` fails the test |
| Versioned named-tensor bundle: name, shape, dtype, byte order, fragments | `TensorBundle`/`NamedTensor`/`TensorFragment`; `test_named_tensor_bundle_is_versioned_and_fully_described`, `test_bundle_round_trips_multiple_named_tensors` |
| Round-trip + compatibility tests in Python and C++ | 29 Python tests; C++ `ctest` 1/1; cross-language byte equality |
| Targeted pytest passes | 29 passed |
| Round-trip + compatibility tests in Python and C++ | 45 Python tests; C++ `ctest` 1/1; cross-language byte equality |
| Targeted pytest passes | 45 passed |
| `compileall packages tests` | exit 0 |
| `git diff --check` | exit 0 |
| Default tests deterministic, download-free, credit-free, GPU-free | Pure in-memory protobuf; no model, no network, no GPU |
| Full deterministic pytest passes, or pre-existing failure recorded | 712 passed, 12 skipped; flaky pre-existing test documented above with clean-tree reproduction |
| Full deterministic pytest passes, or pre-existing failure recorded | Final exact-code run: 728 passed, 12 skipped; earlier sole flaky failure documented with clean-tree reproduction and 3/3 passing retries |
## 5. How the cross-language claim is actually earned