mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-30 19:09:28 +00:00
Second, deeper pass over tools/gateway/hermes_cli plus first pass over the trees wave 1 missed (acp, acp_adapter, skills, computer_use, docker, dashboard, conformance, monitoring, secret_sources, hermes_state, providers). Same rubric as wave 1 (AGENTS.md test policy); security, alternation/caching invariants, issue-number regressions, and E2E kept. Real test-quality fixes found and rooted out along the way: - tests/tools/test_command_guards.py made real auxiliary-LLM HTTPS calls (DEFAULT_CONFIG smart-approval leaked in) — pinned approval mode=manual via autouse fixture: 17.4s → 0.4s. - test_model_switch_custom_providers.py / test_user_providers_model_switch.py silently probed live provider catalogs (~2s/test) — stubbed cached_provider_model_ids/provider_model_ids/fetch_api_models. - test_telegram_noise_filter.py: 15-platform copy-paste matrix over shared gateway.run logic → 3 representative platforms (55s → 3.9s). - test_gateway_shutdown.py: stop()'s 5s interrupt-deadline loop spun on MagicMock agents — interrupt.side_effect now clears _running_agents (22s → 1.0s). - test_gateway_inactivity_timeout.py poll-harness timings shrunk 3-5x (24s → 1.1s); test_mcp_stability.py backoff/SIGTERM-grace sleeps patched (15.4s → 2.5s); test_async_delegation.py negative-drain wait 5s → 0.5s. - test_telegram_init_deadline.py: loop-block margin restored to 1.0s with rationale comment — the watchdog-dump assertion needs the loop blocked well past deadline+grace under parallel load (flaked once in the 40-worker verification run at a 0.2s margin). Verification: full hermetic suite via scripts/run_tests.sh — 2,438 files, 21,718 tests passed, 0 failed, 293.9s wall. Suite totals vs original baseline: 46,820 → 19,757 test functions (−57.8%), wall 583.5s → 293.9s (−50%), subprocess CPU 13,564s → 11,623s.
134 lines
4.4 KiB
Python
134 lines
4.4 KiB
Python
"""Regression test for #40695 (salvage of keystone PR #40782).
|
|
|
|
The Discord gateway heartbeat was stalling because the handoff watcher
|
|
(``GatewayRunner._handoff_watcher``) polled the synchronous, blocking
|
|
SQLite-backed ``SessionDB`` directly on the asyncio event loop every 2s
|
|
('Shard ID None heartbeat blocked for more than N seconds').
|
|
|
|
The fix routes every blocking ``SessionDB`` call in the watcher through the
|
|
``AsyncSessionDB`` facade, which offloads each call via ``asyncio.to_thread`` so
|
|
the SQLite I/O runs on a worker thread and never blocks the event loop / Discord
|
|
heartbeat.
|
|
|
|
These tests assert that behaviour contract. They are mutation-survivable:
|
|
reverting any ``await self._session_db.<call>(...)`` back to a direct synchronous
|
|
call on the loop makes the relevant assertion fail.
|
|
"""
|
|
|
|
import asyncio
|
|
import types
|
|
|
|
import pytest
|
|
|
|
import gateway.run as run
|
|
|
|
|
|
class _RecordingSessionDB:
|
|
"""SessionDB stand-in that records the thread each method runs on.
|
|
|
|
If the watcher calls these methods directly on the event loop (the bug),
|
|
they run on the loop thread. If they are wrapped in ``asyncio.to_thread``
|
|
(the fix), they run on a *different* worker thread.
|
|
"""
|
|
|
|
def __init__(self, loop_thread_ident):
|
|
self._loop_thread_ident = loop_thread_ident
|
|
self.threads = {}
|
|
self.calls = []
|
|
|
|
def _record(self, name):
|
|
import threading
|
|
|
|
self.threads.setdefault(name, []).append(threading.get_ident())
|
|
self.calls.append(name)
|
|
|
|
def ran_off_loop(self, name):
|
|
"""True iff every call to ``name`` ran on a non-loop thread."""
|
|
idents = self.threads.get(name, [])
|
|
return bool(idents) and all(i != self._loop_thread_ident for i in idents)
|
|
|
|
def list_pending_handoffs(self):
|
|
self._record("list_pending_handoffs")
|
|
return [{"id": "sess-1"}]
|
|
|
|
def claim_handoff(self, session_id):
|
|
self._record("claim_handoff")
|
|
return True
|
|
|
|
def complete_handoff(self, session_id):
|
|
self._record("complete_handoff")
|
|
|
|
def fail_handoff(self, session_id, error):
|
|
self._record("fail_handoff")
|
|
|
|
|
|
def _make_fake_runner(session_db, *, fail_process=False):
|
|
"""Build a minimal object that exposes exactly what the loop body touches.
|
|
|
|
The watcher now talks to the SessionDB through the AsyncSessionDB facade,
|
|
so wrap the recording stand-in the same way the gateway does.
|
|
"""
|
|
from hermes_state import AsyncSessionDB
|
|
|
|
fake = types.SimpleNamespace()
|
|
fake._session_db = AsyncSessionDB(session_db)
|
|
# _running yields True for the first loop check, then False so the loop
|
|
# exits after a single tick.
|
|
states = iter([True, False])
|
|
|
|
class _Running:
|
|
def __bool__(_self):
|
|
try:
|
|
return next(states)
|
|
except StopIteration:
|
|
return False
|
|
|
|
fake._running = _Running()
|
|
|
|
async def _process_handoff(row):
|
|
if fail_process:
|
|
raise RuntimeError("boom")
|
|
|
|
fake._process_handoff = _process_handoff
|
|
return fake
|
|
|
|
|
|
async def _run_one_tick(fake, monkeypatch):
|
|
"""Run the watcher for a single tick with sleeps neutralised."""
|
|
|
|
async def _no_sleep(_seconds):
|
|
return None
|
|
|
|
monkeypatch.setattr(run.asyncio, "sleep", _no_sleep)
|
|
# Bind the real (patched) method onto our minimal stand-in.
|
|
coro = run.GatewayRunner._handoff_watcher(fake, interval=0.0)
|
|
await asyncio.wait_for(coro, timeout=5)
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_watcher_wraps_calls_via_asyncio_to_thread(monkeypatch):
|
|
"""Explicitly assert the offload goes through asyncio.to_thread.
|
|
|
|
Patches the AsyncSessionDB facade's ``asyncio.to_thread`` (it lives in
|
|
hermes_state) and records which SessionDB callables were handed to it.
|
|
Mutation-survivable: dropping any await removes its callable from the set.
|
|
"""
|
|
import hermes_state
|
|
|
|
db = _RecordingSessionDB(loop_thread_ident=-1)
|
|
fake = _make_fake_runner(db, fail_process=False)
|
|
|
|
wrapped = []
|
|
real_to_thread = hermes_state.asyncio.to_thread
|
|
|
|
async def _spy_to_thread(func, *args, **kwargs):
|
|
wrapped.append(getattr(func, "__name__", repr(func)))
|
|
return await real_to_thread(func, *args, **kwargs)
|
|
|
|
monkeypatch.setattr(hermes_state.asyncio, "to_thread", _spy_to_thread)
|
|
|
|
await _run_one_tick(fake, monkeypatch)
|
|
|
|
assert "list_pending_handoffs" in wrapped
|
|
assert "claim_handoff" in wrapped
|
|
assert "complete_handoff" in wrapped
|