tractor/tests/test_infected_asyncio.py

982 lines
28 KiB
Python
Raw Permalink Normal View History

2021-11-05 14:42:43 +00:00
'''
The hipster way to force SC onto the stdlib's "async": 'infection mode'.
2021-11-05 14:42:43 +00:00
'''
import asyncio
2021-11-07 22:05:40 +00:00
import builtins
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
from contextlib import ExitStack
from functools import partial
import itertools
2021-11-07 22:05:40 +00:00
import importlib
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
import os
from pathlib import Path
import signal
from typing import (
Callable,
Iterable,
Union,
)
import pytest
2021-11-07 22:05:40 +00:00
import trio
import tractor
from tractor import (
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
current_actor,
Actor,
to_asyncio,
RemoteActorError,
ContextCancelled,
_state,
)
from tractor.trionics import BroadcastReceiver
Add (back) a `tractor._testing` sub-pkg Since importing from our top level `conftest.py` is not scaleable or as "future forward thinking" in terms of: - LoC-wise (it's only one file), - prevents "external" (aka non-test) example scripts from importing content easily, - seemingly(?) can't be used via abs-import if using a `[tool.pytest.ini_options]` in a `pyproject.toml` vs. a `pytest.ini`, see: https://docs.pytest.org/en/8.0.x/reference/customize.html#pyproject-toml) => Go back to having an internal "testing" pkg like `trio` (kinda) does. Deats: - move generic top level helpers into pkg-mod including the new `expect_ctxc()` (which i needed in the advanced faults testing script. - move `@tractor_test` into `._testing.pytest` sub-mod. - adjust all the helper imports to be a `from tractor._testing import <..>` Rework `test_ipc_channel_break_during_stream()` and backing script: - make test(s) pull `debug_mode` from new fixture (which is now controlled manually from `--tpdb` flag) and drop the previous parametrized input. - update logic in ^ test for "which-side-fails" cases to better match recently updated/stricter cancel/failure semantics in terms of `ClosedResouruceError` vs. `EndOfChannel` expectations. - handle `ExceptionGroup`s with expected embedded errors in test. - better pendantics around whether to expect a user simulated KBI. - for `examples/advanced_faults/ipc_failure_during_stream.py` script: - generalize ipc breakage in new `break_ipc()` with support for diff internal `trio` methods and a #TODO for future disti frameworks - only make one sub-actor task break and the other just stream. - use new `._testing.expect_ctxc()` around ctx block. - add a bit of exception handling with `print()`s around ctxc (unused except if 'msg' break method is set) and eoc cases. - don't break parent side ipc in loop any more then once after first break, checked via flag var. - add a `pre_close: bool` flag to control whether `MsgStreama.aclose()` is called *before* any ipc breakage method. Still TODO: - drop `pytest.ini` and add the alt section to `pyproject.py`. -> currently can't get `--rootdir=` opt to work.. not showing in console header. -> ^ also breaks on 'tests' `enable_modules` imports in subactors during discovery tests?
2024-03-12 19:48:20 +00:00
from tractor._testing import expect_ctxc
2021-11-05 14:42:43 +00:00
async def sleep_and_err(
sleep_for: float = 0.1,
# just signature placeholders for compat with
# ``to_asyncio.open_channel_from()``
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
to_trio: trio.MemorySendChannel|None = None,
from_trio: asyncio.Queue|None = None,
):
if to_trio:
to_trio.send_nowait('start')
await asyncio.sleep(sleep_for)
assert 0
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
async def aio_sleep_forever():
2021-11-07 22:05:40 +00:00
await asyncio.sleep(float('inf'))
async def trio_cancels_single_aio_task():
# spawn an ``asyncio`` task to run a func and return result
with trio.move_on_after(.2):
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
await tractor.to_asyncio.run_task(aio_sleep_forever)
def test_trio_cancels_aio_on_actor_side(reg_addr):
'''
Spawn an infected actor that is cancelled by the ``trio`` side
task using std cancel scope apis.
'''
async def main():
async with tractor.open_nursery(
registry_addrs=[reg_addr]
) as n:
await n.run_in_actor(
trio_cancels_single_aio_task,
infect_asyncio=True,
)
trio.run(main)
2021-11-07 22:05:40 +00:00
async def asyncio_actor(
target: str,
expect_err: Exception|None = None
2021-11-07 22:05:40 +00:00
) -> None:
# ensure internal runtime state is consistent
actor: Actor = tractor.current_actor()
assert (
actor.is_infected_aio()
and
actor._infected_aio
and
_state._runtime_vars['_is_infected_aio']
)
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
target: Callable = globals()[target]
2021-11-07 22:05:40 +00:00
if '.' in expect_err:
modpath, _, name = expect_err.rpartition('.')
mod = importlib.import_module(modpath)
error_type = getattr(mod, name)
else: # toplevel builtin error type
error_type = builtins.__dict__.get(expect_err)
2021-11-07 22:05:40 +00:00
try:
# spawn an ``asyncio`` task to run a func and return result
await tractor.to_asyncio.run_task(target)
except BaseException as err:
2021-11-07 22:05:40 +00:00
if expect_err:
assert isinstance(err, error_type)
2021-11-07 22:05:40 +00:00
raise
def test_aio_simple_error(reg_addr):
2021-11-07 22:05:40 +00:00
'''
Verify a simple remote asyncio error propagates back through trio
to the parent actor.
2021-11-07 22:05:40 +00:00
'''
async def main():
2021-11-07 22:05:40 +00:00
async with tractor.open_nursery(
registry_addrs=[reg_addr]
2021-11-07 22:05:40 +00:00
) as n:
await n.run_in_actor(
asyncio_actor,
target='sleep_and_err',
expect_err='AssertionError',
infect_asyncio=True,
)
with pytest.raises(
expected_exception=(RemoteActorError, ExceptionGroup),
) as excinfo:
2021-11-07 22:05:40 +00:00
trio.run(main)
2021-11-05 14:42:43 +00:00
2021-11-07 22:05:40 +00:00
err = excinfo.value
# might get multiple `trio.Cancelled`s as well inside an inception
if isinstance(err, ExceptionGroup):
err = next(itertools.dropwhile(
lambda exc: not isinstance(exc, tractor.RemoteActorError),
err.exceptions
))
assert err
2021-11-07 22:05:40 +00:00
assert isinstance(err, RemoteActorError)
assert err.boxed_type is AssertionError
2021-11-05 14:42:43 +00:00
2021-11-07 22:05:40 +00:00
def test_tractor_cancels_aio(reg_addr):
2021-11-07 22:05:40 +00:00
'''
Verify we can cancel a spawned asyncio task gracefully.
'''
async def main():
async with tractor.open_nursery() as n:
portal = await n.run_in_actor(
asyncio_actor,
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
target='aio_sleep_forever',
expect_err='trio.Cancelled',
2021-11-07 22:05:40 +00:00
infect_asyncio=True,
)
# cancel the entire remote runtime
2021-11-07 22:05:40 +00:00
await portal.cancel_actor()
trio.run(main)
2021-11-05 14:42:43 +00:00
def test_trio_cancels_aio(reg_addr):
'''
Much like the above test with ``tractor.Portal.cancel_actor()``
except we just use a standard ``trio`` cancellation api.
'''
async def main():
with trio.move_on_after(1):
# cancel the nursery shortly after boot
async with tractor.open_nursery() as n:
await n.run_in_actor(
asyncio_actor,
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
target='aio_sleep_forever',
expect_err='trio.Cancelled',
infect_asyncio=True,
)
trio.run(main)
@tractor.context
async def trio_ctx(
ctx: tractor.Context,
):
await ctx.started('start')
# this will block until the ``asyncio`` task sends a "first"
# message.
2022-07-15 00:44:10 +00:00
with trio.fail_after(2):
async with (
trio.open_nursery() as n,
tractor.to_asyncio.open_channel_from(
sleep_and_err,
) as (first, chan),
):
assert first == 'start'
# spawn another asyncio task for the cuck of it.
n.start_soon(
tractor.to_asyncio.run_task,
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
aio_sleep_forever,
)
await trio.sleep_forever()
@pytest.mark.parametrize(
'parent_cancels',
['context', 'actor', False],
ids='parent_actor_cancels_child={}'.format
)
def test_context_spawns_aio_task_that_errors(
reg_addr,
parent_cancels: bool,
):
'''
Verify that spawning a task via an intertask channel ctx mngr that
errors correctly propagates the error back from the `asyncio`-side
2022-07-15 15:18:48 +00:00
task.
'''
async def main():
2022-10-11 17:35:55 +00:00
with trio.fail_after(2):
async with tractor.open_nursery() as n:
p = await n.start_actor(
'aio_daemon',
enable_modules=[__name__],
infect_asyncio=True,
# debug_mode=True,
loglevel='cancel',
)
async with (
expect_ctxc(
yay=parent_cancels == 'actor',
),
p.open_context(
trio_ctx,
) as (ctx, first),
):
2022-10-11 17:35:55 +00:00
assert first == 'start'
if parent_cancels == 'actor':
2022-10-11 17:35:55 +00:00
await p.cancel_actor()
elif parent_cancels == 'context':
await ctx.cancel()
else:
await trio.sleep_forever()
async with expect_ctxc(
yay=parent_cancels == 'actor',
):
await ctx.result()
if parent_cancels == 'context':
# to tear down sub-acor
await p.cancel_actor()
return ctx.outcome
if parent_cancels:
# bc the parent made the cancel request,
# the error is not raised locally but instead
# the context is exited silently
res = trio.run(main)
assert isinstance(res, ContextCancelled)
assert 'root' in res.canceller[0]
else:
expect = RemoteActorError
with pytest.raises(expect) as excinfo:
trio.run(main)
err = excinfo.value
assert isinstance(err, expect)
assert err.boxed_type is AssertionError
async def aio_cancel():
''''
Cancel urself boi.
'''
await asyncio.sleep(0.5)
# cancel and enter sleep
task = asyncio.current_task()
task.cancel()
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
await aio_sleep_forever()
def test_aio_cancelled_from_aio_causes_trio_cancelled(reg_addr):
'''
When the `asyncio.Task` cancels itself the `trio` side cshould
also cancel and teardown and relay the cancellation cross-process
to the caller (parent).
'''
async def main():
an: tractor.ActorNursery
async with tractor.open_nursery() as an:
p: tractor.Portal = await an.run_in_actor(
asyncio_actor,
target='aio_cancel',
expect_err='tractor.to_asyncio.AsyncioCancelled',
infect_asyncio=True,
)
# NOTE: normally the `an.__aexit__()` waits on the
# portal's result but we do it explicitly here
# to avoid indent levels.
with trio.fail_after(1):
await p.wait_for_result()
with pytest.raises(
expected_exception=(RemoteActorError, ExceptionGroup),
) as excinfo:
trio.run(main)
2021-11-05 14:42:43 +00:00
# might get multiple `trio.Cancelled`s as well inside an inception
err: RemoteActorError|ExceptionGroup = excinfo.value
if isinstance(err, ExceptionGroup):
err = next(itertools.dropwhile(
lambda exc: not isinstance(exc, tractor.RemoteActorError),
err.exceptions
))
assert err
# relayed boxed error should be our `trio`-task's
# cancel-signal-proxy-equivalent of `asyncio.CancelledError`.
assert err.boxed_type == to_asyncio.AsyncioCancelled
2021-11-05 14:42:43 +00:00
# TODO: verify open_channel_from will fail on this..
async def no_to_trio_in_args():
pass
async def push_from_aio_task(
sequence: Iterable,
to_trio: trio.abc.SendChannel,
expect_cancel: False,
fail_early: bool,
) -> None:
try:
# sync caller ctx manager
to_trio.send_nowait(True)
for i in sequence:
print(f'asyncio sending {i}')
to_trio.send_nowait(i)
await asyncio.sleep(0.001)
if i == 50 and fail_early:
raise Exception
print('asyncio streamer complete!')
except asyncio.CancelledError:
if not expect_cancel:
pytest.fail("aio task was cancelled unexpectedly")
raise
else:
if expect_cancel:
pytest.fail("aio task wasn't cancelled as expected!?")
async def stream_from_aio(
exit_early: bool = False,
raise_err: bool = False,
aio_raise_err: bool = False,
fan_out: bool = False,
) -> None:
seq = range(100)
expect = list(seq)
try:
pulled = []
async with to_asyncio.open_channel_from(
push_from_aio_task,
sequence=seq,
expect_cancel=raise_err or exit_early,
fail_early=aio_raise_err,
) as (first, chan):
assert first is True
async def consume(
chan: Union[
to_asyncio.LinkedTaskChannel,
BroadcastReceiver,
],
):
async for value in chan:
print(f'trio received {value}')
pulled.append(value)
if value == 50:
if raise_err:
raise Exception
elif exit_early:
break
if fan_out:
# start second task that get's the same stream value set.
async with (
# NOTE: this has to come first to avoid
# the channel being closed before the nursery
# tasks are joined..
chan.subscribe() as br,
trio.open_nursery() as n,
):
n.start_soon(consume, br)
await consume(chan)
else:
await consume(chan)
finally:
if (
not raise_err and
not exit_early and
not aio_raise_err
):
if fan_out:
# we get double the pulled values in the
# ``.subscribe()`` fan out case.
doubled = list(itertools.chain(*zip(expect, expect)))
2022-04-12 20:13:33 +00:00
expect = doubled[:len(pulled)]
assert list(sorted(pulled)) == expect
else:
assert pulled == expect
else:
2022-04-12 20:13:33 +00:00
assert not fan_out
assert pulled == expect[:51]
print('trio guest mode task completed!')
2021-11-05 14:42:43 +00:00
@pytest.mark.parametrize(
'fan_out', [False, True],
ids='fan_out_w_chan_subscribe={}'.format
)
def test_basic_interloop_channel_stream(reg_addr, fan_out):
async def main():
async with tractor.open_nursery() as n:
portal = await n.run_in_actor(
stream_from_aio,
infect_asyncio=True,
fan_out=fan_out,
)
# should raise RAE diectly
await portal.result()
trio.run(main)
# TODO: parametrize the above test and avoid the duplication here?
def test_trio_error_cancels_intertask_chan(reg_addr):
async def main():
async with tractor.open_nursery() as n:
portal = await n.run_in_actor(
stream_from_aio,
raise_err=True,
infect_asyncio=True,
)
# should trigger remote actor error
await portal.result()
with pytest.raises(RemoteActorError) as excinfo:
trio.run(main)
2021-11-05 14:42:43 +00:00
# ensure boxed error type
excinfo.value.boxed_type is Exception
2021-11-05 14:42:43 +00:00
def test_trio_closes_early_and_channel_exits(reg_addr):
async def main():
async with tractor.open_nursery() as n:
portal = await n.run_in_actor(
stream_from_aio,
exit_early=True,
infect_asyncio=True,
)
# should raise RAE diectly
await portal.result()
# should be a quiet exit on a simple channel exit
trio.run(main)
def test_aio_errors_and_channel_propagates_and_closes(reg_addr):
async def main():
async with tractor.open_nursery() as n:
portal = await n.run_in_actor(
stream_from_aio,
aio_raise_err=True,
infect_asyncio=True,
)
# should trigger RAE directly, not an eg.
await portal.result()
with pytest.raises(
# NOTE: bc we directly wait on `Portal.result()` instead
# of capturing it inside the `ActorNursery` machinery.
expected_exception=RemoteActorError,
) as excinfo:
trio.run(main)
excinfo.value.boxed_type is Exception
2021-11-05 14:42:43 +00:00
async def aio_echo_server(
to_trio: trio.MemorySendChannel,
from_trio: asyncio.Queue,
) -> None:
2021-11-25 22:10:22 +00:00
to_trio.send_nowait('start')
2021-11-25 22:10:22 +00:00
while True:
msg = await from_trio.get()
2021-11-25 22:10:22 +00:00
# echo the msg back
to_trio.send_nowait(msg)
2021-11-25 22:10:22 +00:00
# if we get the terminate sentinel
# break the echo loop
if msg is None:
print('breaking aio echo loop')
break
2021-11-25 22:10:22 +00:00
print('exiting asyncio task')
2021-11-25 22:10:22 +00:00
@tractor.context
async def trio_to_aio_echo_server(
ctx: tractor.Context|None,
):
2021-11-25 22:10:22 +00:00
async with to_asyncio.open_channel_from(
aio_echo_server,
) as (first, chan):
assert first == 'start'
2021-11-25 22:10:22 +00:00
await ctx.started(first)
async with ctx.open_stream() as stream:
async for msg in stream:
print(f'asyncio echoing {msg}')
await chan.send(msg)
out = await chan.receive()
# echo back to parent actor-task
await stream.send(out)
if out is None:
try:
out = await chan.receive()
except trio.EndOfChannel:
break
else:
raise RuntimeError('aio channel never stopped?')
2021-11-28 02:55:04 +00:00
@pytest.mark.parametrize(
'raise_error_mid_stream',
[False, Exception, KeyboardInterrupt],
ids='raise_error={}'.format,
)
def test_echoserver_detailed_mechanics(
reg_addr,
2021-11-28 02:55:04 +00:00
raise_error_mid_stream,
):
2021-11-25 22:10:22 +00:00
async def main():
async with tractor.open_nursery() as n:
p = await n.start_actor(
'aio_server',
enable_modules=[__name__],
infect_asyncio=True,
)
async with p.open_context(
trio_to_aio_echo_server,
) as (ctx, first):
assert first == 'start'
async with ctx.open_stream() as stream:
for i in range(100):
await stream.send(i)
out = await stream.receive()
assert i == out
2021-11-28 02:55:04 +00:00
if raise_error_mid_stream and i == 50:
raise raise_error_mid_stream
2021-11-25 22:10:22 +00:00
# send terminate msg
await stream.send(None)
out = await stream.receive()
assert out is None
if out is None:
# ensure the stream is stopped
# with trio.fail_after(0.1):
try:
await stream.receive()
except trio.EndOfChannel:
pass
else:
pytest.fail(
'stream not stopped after sentinel ?!'
)
2021-11-25 22:10:22 +00:00
# TODO: the case where this blocks and
# is cancelled by kbi or out of task cancellation
await p.cancel_actor()
2021-11-28 02:55:04 +00:00
if raise_error_mid_stream:
with pytest.raises(raise_error_mid_stream):
trio.run(main)
else:
trio.run(main)
@pytest.mark.parametrize(
'raise_error_mid_stream',
[
False,
Exception,
KeyboardInterrupt,
],
ids='raise_error={}'.format,
)
def test_infected_root_actor(
raise_error_mid_stream: bool|Exception,
# conftest wide
loglevel: str,
debug_mode: bool,
):
'''
Verify you can run the `tractor` runtime with `Actor.is_infected_aio() == True`
in the root actor.
'''
async def _trio_main():
first: str
chan: to_asyncio.LinkedTaskChannel
async with (
tractor.open_root_actor(
debug_mode=debug_mode,
loglevel=loglevel,
),
to_asyncio.open_channel_from(
aio_echo_server,
) as (first, chan),
):
assert first == 'start'
for i in range(1000):
await chan.send(i)
out = await chan.receive()
assert out == i
print(f'asyncio echoing {i}')
if raise_error_mid_stream and i == 500:
raise raise_error_mid_stream
if out is None:
try:
out = await chan.receive()
except trio.EndOfChannel:
break
else:
raise RuntimeError('aio channel never stopped?')
if raise_error_mid_stream:
with pytest.raises(raise_error_mid_stream):
tractor.to_asyncio.run_as_asyncio_guest(
trio_main=_trio_main,
)
else:
tractor.to_asyncio.run_as_asyncio_guest(
trio_main=_trio_main,
)
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
@tractor.context
async def manage_file(
ctx: tractor.Context,
tmp_path_str: str,
send_sigint_to: str,
trio_side_is_shielded: bool = True,
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
bg_aio_task: bool = False,
):
'''
Start an `asyncio` task that just sleeps after registering a context
with `Actor.lifetime_stack`. Trigger a SIGINT to kill the actor tree
and ensure the stack is closed in the infected mode child.
To verify the teardown state just write a tmpfile to the `testdir`
and delete it on actor close.
'''
tmp_path: Path = Path(tmp_path_str)
tmp_file: Path = tmp_path / f'{" ".join(ctx._actor.uid)}.file'
# create a the tmp file and tell the parent where it's at
assert not tmp_file.is_file()
tmp_file.touch()
stack: ExitStack = current_actor().lifetime_stack
stack.callback(tmp_file.unlink)
await ctx.started((
str(tmp_file),
os.getpid(),
))
# expect to be cancelled from here!
try:
# NOTE: turns out you don't even need to sched an aio task
# since the original issue, even though seemingly was due to
# the guest-run being abandoned + a `._debug.pause()` inside
# `._runtime._async_main()` (which was originally trying to
# debug the `.lifetime_stack` not closing), IS NOT actually
# the core issue?
#
# further notes:
#
# - `trio` only issues the " RuntimeWarning: Trio guest run
# got abandoned without properly finishing... weird stuff
# might happen" IFF you DO run a asyncio task here, BUT
# - the original issue of the `.lifetime_stack` not closing
# will still happen even if you don't run an `asyncio` task
# here even though the "abandon" messgage won't be shown..
#
# => ????? honestly i'm lost but it seems to be some issue
# with `asyncio` and SIGINT..
#
# honestly, this REALLY reminds me why i haven't used
# `asyncio` by choice in years.. XD
#
async with trio.open_nursery() as tn:
if bg_aio_task:
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
tn.start_soon(
tractor.to_asyncio.run_task,
aio_sleep_forever,
)
# XXX don't-need/doesn't-make-a-diff right
# since we're already doing it from parent?
# if send_sigint_to == 'child':
# os.kill(
# os.getpid(),
# signal.SIGINT,
# )
# XXX spend a half sec doing shielded checkpointing to
# ensure that despite the `trio`-side task ignoring the
# SIGINT, the `asyncio` side won't abandon the guest-run!
if trio_side_is_shielded:
with trio.CancelScope(shield=True):
for i in range(5):
await trio.sleep(0.1)
await trio.sleep_forever()
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
# signalled manually at the OS level (aka KBI) by the parent actor.
except KeyboardInterrupt:
print('child raised KBI..')
assert tmp_file.exists()
raise
raise RuntimeError('shoulda received a KBI?')
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
@pytest.mark.parametrize(
'trio_side_is_shielded',
[
False,
True,
],
ids=[
'trio_side_no_shielding',
'trio_side_does_shielded_work',
],
)
@pytest.mark.parametrize(
'send_sigint_to',
[
'child',
'parent',
],
ids='send_SIGINT_to={}'.format,
)
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
@pytest.mark.parametrize(
'bg_aio_task',
[
False,
# NOTE: (and see notes in `manage_file()` above as well) if
# we FOR SURE SPAWN AN AIO TASK in the child it seems the
# "silent-abandon" case (as is described in detail in
# `to_asyncio.run_as_asyncio_guest()`) does not happen and
# `asyncio`'s loop will at least abandon the `trio` side
# loudly? .. prolly the state-spot to start looking for
# a soln that results in NO ABANDONMENT.. XD
True,
],
ids=[
'bg_aio_task',
'just_trio_slee',
],
)
@pytest.mark.parametrize(
'wait_for_ctx',
[
False,
True,
],
ids=[
'raise_KBI_in_rent',
'wait_for_ctx',
],
)
def test_sigint_closes_lifetime_stack(
tmp_path: Path,
wait_for_ctx: bool,
bg_aio_task: bool,
trio_side_is_shielded: bool,
debug_mode: bool,
send_sigint_to: str,
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
):
'''
Ensure that an infected child can use the `Actor.lifetime_stack`
to make a file on boot and it's automatically cleaned up by the
actor-lifetime-linked exit stack closure.
'''
async def main():
try:
an: tractor.ActorNursery
async with tractor.open_nursery(
debug_mode=debug_mode,
) as an:
p: tractor.Portal = await an.start_actor(
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
'file_mngr',
enable_modules=[__name__],
infect_asyncio=True,
)
async with p.open_context(
manage_file,
tmp_path_str=str(tmp_path),
send_sigint_to=send_sigint_to,
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
bg_aio_task=bg_aio_task,
trio_side_is_shielded=trio_side_is_shielded,
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
) as (ctx, first):
path_str, cpid = first
tmp_file: Path = Path(path_str)
assert tmp_file.exists()
# XXX originally to simulate what (hopefully)
# the below now triggers.. had to manually
# trigger a SIGINT from a ctl-c in the root.
# await trio.sleep_forever()
# XXX NOTE XXX signal infected-`asyncio` child to
# OS-cancel with SIGINT; this should trigger the
# bad `asyncio` cancel behaviour that can cause
# a guest-run abandon as was seen causing
# shm-buffer leaks in `piker`'s live quote stream
# susbys!
#
await trio.sleep(.2)
pid: int = (
cpid if send_sigint_to == 'child'
else os.getpid()
)
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
os.kill(
pid,
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
signal.SIGINT,
)
# XXX CASE 1: without the bug fixed, in
# the non-KBI-raised-in-parent case, this
# timeout should trigger!
if wait_for_ctx:
print('waiting for ctx outcome in parent..')
try:
with trio.fail_after(1):
Hack `asyncio` to not abandon a guest-mode run? Took me a while to figure out what the heck was going on but, turns out `asyncio` changed their SIGINT handling in 3.11 as per: https://docs.python.org/3/library/asyncio-runner.html#handling-keyboard-interruption I'm not entirely sure if it's the 3.11 changes or possibly wtv further updates were made in 3.12 but more or less due to the way our current main task was written the `trio` guest-run was getting abandoned on SIGINTs sent from the OS to the infected child proc.. Note that much of the bug and soln cases are layed out in very detailed comment-notes both in the new test and `run_as_asyncio_guest()`, right above the final "fix" lines. Add new `test_infected_aio.test_sigint_closes_lifetime_stack()` test suite which reliably triggers all abandonment issues with multiple cases of different parent behaviour post-sending-SIGINT-to-child: 1. briefly sleep then raise a KBI in the parent which was originally demonstrating the file leak not being cleaned up by `Actor.lifetime_stack.close()` and simulates a ctl-c from the console (relayed in tandem by the OS to the parent and child processes). 2. do `Context.wait_for_result()` on the child context which would hang and timeout since the actor runtime would never complete and thus never relay a `ContextCancelled`. 3. both with and without running a `asyncio` task in the `manage_file` child actor; originally it seemed that with an aio task scheduled in the child actor the guest-run abandonment always was the "loud" case where there seemed to be some actor teardown but with tbs from python failing to gracefully exit the `trio` runtime.. The (seemingly working) "fix" required 2 lines of code to be run inside a `asyncio.CancelledError` handler around the call to `await trio_done_fut`: - `Actor.cancel_soon()` which schedules the actor runtime to cancel on the next `trio` runner cycle and results in a "self cancellation" of the actor. - "pumping the `asyncio` event loop" with a non-0 `.sleep(0.1)` XD |_ seems that a "shielded" pump with some actual `delay: float >= 0` did the trick to get `asyncio` to allow the `trio` runner/loop to fully complete its guest-run without abandonment. Other supporting changes: - move `._exceptions.AsyncioCancelled`, our renamed `asyncio.CancelledError` error-sub-type-wrapper, to `.to_asyncio` and make it derive from `CancelledError` so as to be sure when raised by our `asyncio` x-> `trio` exception relay machinery that `asyncio` is getting the specific type it expects during cancellation. - do "summary status" style logging in `run_as_asyncio_guest()` wherein we compile the eventual `startup_msg: str` emitted just before waiting on the `trio_done_fut`. - shield-wait with `out: Outcome = await asyncio.shield(trio_done_fut)` even though it seems to do nothing in the SIGINT handling case..(I presume it might help avoid abandonment in a `asyncio.Task.cancel()` case maybe?)
2024-06-24 20:10:23 +00:00
await ctx.wait_for_result()
except tractor.ContextCancelled as ctxc:
assert ctxc.canceller == ctx.chan.uid
raise
# XXX CASE 2: this seems to be the source of the
# original issue which exhibited BEFORE we put
# a `Actor.cancel_soon()` inside
# `run_as_asyncio_guest()`..
else:
raise KeyboardInterrupt
pytest.fail('should have raised some kinda error?!?')
except (
KeyboardInterrupt,
ContextCancelled,
):
# XXX CASE 2: without the bug fixed, in the
# KBI-raised-in-parent case, the actor teardown should
# never get run (silently abaondoned by `asyncio`..) and
# thus the file should leak!
assert not tmp_file.exists()
assert ctx.maybe_error
trio.run(main)
# TODO: debug_mode tests once we get support for `asyncio`!
#
# -[ ] need tests to wrap both scripts:
# - [ ] infected_asyncio_echo_server.py
# - [ ] debugging/asyncio_bp.py
# -[ ] consider moving ^ (some of) these ^ to `test_debugger`?
#
# -[ ] missing impl outstanding includes:
# - [x] for sync pauses we need to ensure we open yet another
# `greenback` portal in the asyncio task
# => completed using `.bestow_portal(task)` inside
# `.to_asyncio._run_asyncio_task()` right?
# -[ ] translation func to get from `asyncio` task calling to
# `._debug.wait_for_parent_stdin_hijack()` which does root
# call to do TTY locking.
#
def test_sync_breakpoint():
'''
Verify we can do sync-func/code breakpointing using the
`breakpoint()` builtin inside infected mode actors.
'''
pytest.xfail('This support is not implemented yet!')
def test_debug_mode_crash_handling():
'''
Verify mult-actor crash handling works with a combo of infected-`asyncio`-mode
and normal `trio` actors despite nested process trees.
'''
pytest.xfail('This support is not implemented yet!')