mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-30 19:09:28 +00:00
39975613b1
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.
112 lines
3.9 KiB
Python
112 lines
3.9 KiB
Python
"""Tests for #60432: the gateway shutdown drain was structurally blind to
|
|
in-flight cron work. Cron jobs run through cron/scheduler.py's own thread
|
|
pool, entirely outside ``GatewayRunner._running_agents`` -- the dict every
|
|
other active-work check on this class reads. A shutdown (``/update``,
|
|
``/restart``, SIGUSR1 -- they all funnel through the same ``stop()``) could
|
|
report ``active_at_start=0`` and immediately kill tool subprocesses while a
|
|
cron job's terminal command was still running.
|
|
|
|
These tests cover the gateway side of the fix:
|
|
- _active_cron_job_count() reads cron.scheduler's in-flight job set
|
|
- _drain_active_agents() waits for cron work the same way it already
|
|
waits for chat sessions
|
|
- the final tool-subprocess kill marks any still-in-flight cron job
|
|
interrupted
|
|
|
|
See tests/cron/test_shutdown_interrupt.py for the cron-side primitives
|
|
this relies on (get_running_job_ids, mark_running_jobs_interrupted).
|
|
"""
|
|
|
|
import asyncio
|
|
from unittest.mock import MagicMock, patch
|
|
|
|
import pytest
|
|
|
|
from tests.gateway.restart_test_helpers import make_restart_runner
|
|
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def _reset_cron_running_set():
|
|
import cron.scheduler as sched
|
|
|
|
sched._running_job_ids.clear()
|
|
sched._interrupted_job_ids.clear()
|
|
yield
|
|
sched._running_job_ids.clear()
|
|
sched._interrupted_job_ids.clear()
|
|
|
|
|
|
def _make_async_noop():
|
|
async def _noop(*args, **kwargs):
|
|
return None
|
|
|
|
return _noop
|
|
|
|
|
|
class TestActiveCronJobCount:
|
|
def test_zero_when_no_cron_jobs_running(self):
|
|
runner, _adapter = make_restart_runner()
|
|
assert runner._active_cron_job_count() == 0
|
|
|
|
|
|
class TestDrainWaitsForCronWork:
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_drain_waits_for_in_flight_cron_job(self):
|
|
"""Before this fix, a cron-only workload made active_at_start=0
|
|
and the drain returned instantly -- this is the exact repro from
|
|
the issue (a `sleep 1800` cron job in flight during /update)."""
|
|
import cron.scheduler as sched
|
|
|
|
runner, _adapter = make_restart_runner()
|
|
sched._running_job_ids.add("job-1")
|
|
|
|
async def finish_job():
|
|
await asyncio.sleep(0.12)
|
|
sched._running_job_ids.discard("job-1")
|
|
|
|
task = asyncio.create_task(finish_job())
|
|
_snapshot, timed_out = await runner._drain_active_agents(2.0)
|
|
await task
|
|
|
|
assert timed_out is False, (
|
|
"drain must wait for the cron job to finish, not report "
|
|
"active_at_start=0 and return instantly"
|
|
)
|
|
|
|
|
|
class TestKillToolSubprocessesMarksCronInterrupted:
|
|
@pytest.mark.asyncio
|
|
async def test_in_flight_cron_job_marked_interrupted_on_forced_kill(self, monkeypatch):
|
|
import cron.scheduler as sched
|
|
import tools.process_registry as _pr
|
|
import tools.terminal_tool as _tt
|
|
import tools.browser_tool as _bt
|
|
|
|
runner, adapter = make_restart_runner()
|
|
runner._restart_drain_timeout = 0.01 # force the timeout path
|
|
adapter.disconnect = _make_async_noop()
|
|
|
|
sched._running_job_ids.add("job-1")
|
|
|
|
monkeypatch.setattr(_pr.process_registry, "kill_all", lambda task_id=None: 1)
|
|
monkeypatch.setattr(_tt, "cleanup_all_environments", lambda: None)
|
|
monkeypatch.setattr(_bt, "cleanup_all_browsers", lambda: None)
|
|
|
|
marked_calls = []
|
|
real_mark = sched.mark_running_jobs_interrupted
|
|
|
|
def _spy(reason):
|
|
result = real_mark(reason)
|
|
marked_calls.append((reason, result))
|
|
return result
|
|
|
|
monkeypatch.setattr(sched, "mark_running_jobs_interrupted", _spy)
|
|
|
|
with patch("gateway.status.remove_pid_file"), patch("gateway.status.write_runtime_status"), \
|
|
patch("cron.scheduler.mark_job_run"):
|
|
await runner.stop()
|
|
|
|
assert marked_calls, "mark_running_jobs_interrupted was never called during shutdown"
|
|
assert any(result == ["job-1"] for _reason, result in marked_calls)
|
|
|