forked from goodboy/tractor
1
0
Fork 0

WIP rework trio spanwer to include cancellation logic; not correct yet..

zombie_lord_infinite
Tyler Goodlet 2021-10-08 18:14:44 -04:00
parent c02a493d8c
commit 64ebb2aff4
1 changed files with 341 additions and 162 deletions

View File

@ -1,6 +1,7 @@
"""
Machinery for actor process spawning using multiple backends.
"""
from __future__ import annotations
import sys
import multiprocessing as mp
import platform
@ -8,7 +9,6 @@ from typing import Any, Dict, Optional
import trio
from trio_typing import TaskStatus
from async_generator import asynccontextmanager
try:
from multiprocessing import semaphore_tracker # type: ignore
@ -32,6 +32,7 @@ from ._portal import Portal
from ._actor import Actor
from ._entry import _mp_main
from ._exceptions import ActorFailure
from ._debug import maybe_wait_for_debugger
log = get_logger('tractor')
@ -90,234 +91,409 @@ def try_set_start_method(name: str) -> Optional[mp.context.BaseContext]:
return _ctx
async def exhaust_portal(
portal: Portal,
actor: Actor
) -> Any:
"""Pull final result from portal (assuming it has one).
# async def exhaust_portal(
# portal: Portal,
# actor: Actor
# ) -> Any:
# """Pull final result from portal (assuming it has one).
If the main task is an async generator do our best to consume
what's left of it.
"""
try:
log.debug(f"Waiting on final result from {actor.uid}")
# If the main task is an async generator do our best to consume
# what's left of it.
# """
# try:
# log.debug(f"Waiting on final result from {actor.uid}")
# XXX: streams should never be reaped here since they should
# always be established and shutdown using a context manager api
final = await portal.result()
# # XXX: streams should never be reaped here since they should
# # always be established and shutdown using a context manager api
# final = await portal.result()
except (Exception, trio.MultiError) as err:
# we reraise in the parent task via a ``trio.MultiError``
return err
except trio.Cancelled as err:
# lol, of course we need this too ;P
# TODO: merge with above?
log.warning(f"Cancelled result waiter for {portal.actor.uid}")
return err
else:
log.debug(f"Returning final result: {final}")
return final
# except (Exception, trio.MultiError) as err:
# # we reraise in the parent task via a ``trio.MultiError``
# return err
# except trio.Cancelled as err:
# # lol, of course we need this too ;P
# # TODO: merge with above?
# log.warning(f"Cancelled result waiter for {portal.actor.uid}")
# return err
# else:
# log.debug(f"Returning final result: {final}")
# return final
async def cancel_on_completion(
async def result_from_portal(
portal: Portal,
actor: Actor,
errors: Dict[Tuple[str, str], Exception],
cancel_on_result: bool = False,
task_status: TaskStatus[trio.CancelScope] = trio.TASK_STATUS_IGNORED,
) -> None:
"""Cancel actor gracefully once it's "main" portal's
"""
Cancel actor gracefully once it's "main" portal's
result arrives.
Should only be called for actors spawned with `run_in_actor()`.
"""
with trio.CancelScope() as cs:
"""
__tracebackhide__ = True
# cancel control is explicityl done by the caller
with trio.CancelScope() as cs:
task_status.started(cs)
# if this call errors we store the exception for later
# in ``errors`` which will be reraised inside
# a MultiError and we still send out a cancel request
result = await exhaust_portal(portal, actor)
if isinstance(result, Exception):
errors[actor.uid] = result
log.warning(
f"Cancelling {portal.channel.uid} after error {result}"
)
# result = await exhaust_portal(portal, actor)
try:
log.debug(f"Waiting on final result from {actor.uid}")
else:
log.runtime(
f"Cancelling {portal.channel.uid} gracefully "
f"after result {result}")
# XXX: streams should never be reaped here since they should
# always be established and shutdown using a context manager api
result = await portal.result()
log.debug(f"Returning final result: {result}")
# cancel the process now that we have a final result
await portal.cancel_actor()
except (Exception, trio.MultiError) as err:
# we reraise in the parent task via a ``trio.MultiError``
result = err
errors[actor.uid] = err
# raise
except trio.Cancelled as err:
# lol, of course we need this too ;P
# TODO: merge with above?
log.warning(f"Cancelled result waiter for {portal.channel.uid}")
result = err
# errors[actor.uid] = err
# raise
if cancel_on_result:
if isinstance(result, Exception):
# errors[actor.uid] = result
log.warning(
f"Cancelling {portal.channel.uid} after error {result}"
)
raise result
else:
log.runtime(
f"Cancelling {portal.channel.uid} gracefully "
f"after result {result}")
# an actor that was `.run_in_actor()` executes a single task
# and delivers the result, then we cancel it.
# TODO: likely in the future we should just implement this using
# the new `open_context()` IPC api, since it's the more general
# api and can represent this form.
# XXX: do we need this?
# await maybe_wait_for_debugger()
await portal.cancel_actor()
return result
async def do_hard_kill(
proc: trio.Process,
timeout: float,
) -> None:
'''
Hard kill a process with timeout.
'''
log.debug(f"Hard killing {proc}")
# NOTE: this timeout used to do nothing since we were shielding
# the ``.wait()`` inside ``new_proc()`` which will pretty much
# never release until the process exits, now it acts as
# a hard-kill time ultimatum.
with trio.move_on_after(3) as cs:
with trio.move_on_after(timeout) as cs:
# NOTE: This ``__aexit__()`` shields internally.
async with proc: # calls ``trio.Process.aclose()``
# NOTE: This ``__aexit__()`` shields internally and originally
# would tear down stdstreams via ``trio.Process.aclose()``.
async with proc:
log.debug(f"Terminating {proc}")
# proc.terminate()
if cs.cancelled_caught:
# XXX: should pretty much never get here unless we have
# to move the bits from ``proc.__aexit__()`` out and
# into here.
log.critical(f"HARD KILLING {proc}")
log.critical(f"{timeout} timeout, HARD KILLING {proc}")
proc.kill()
@asynccontextmanager
async def spawn_subactor(
subactor: 'Actor',
parent_addr: Tuple[str, int],
infect_asyncio: bool,
):
spawn_cmd = [
sys.executable,
"-m",
# Hardcode this (instead of using ``_child.__name__`` to avoid a
# double import warning: https://stackoverflow.com/a/45070583
"tractor._child",
# We provide the child's unique identifier on this exec/spawn
# line for debugging purposes when viewing the process tree from
# the OS; it otherwise can be passed via the parent channel if
# we prefer in the future (for privacy).
"--uid",
str(subactor.uid),
# Address the child must connect to on startup
"--parent_addr",
str(parent_addr)
]
async def reap_proc(
if subactor.loglevel:
spawn_cmd += [
"--loglevel",
subactor.loglevel
]
proc: trio.Process,
terminate_after: float = float('inf'),
hard_kill_after: int = 0.1,
# Tell child to run in guest mode on top of ``asyncio`` loop
if infect_asyncio:
spawn_cmd.append("--asyncio")
) -> None:
with trio.move_on_after(terminate_after) as cs:
# Wait for proc termination but **dont' yet** do
# any out-of-ipc-land termination / process
# killing. This is a "light" (cancellable) join,
# the hard join is below after timeout
await proc.wait()
proc = await trio.open_process(spawn_cmd)
try:
yield proc
if cs.cancelled_caught and terminate_after is not float('inf'):
# Always "hard" join lingering sub procs since no
# actor zombies are allowed!
log.warning(
# f'Failed to gracefully terminate {subactor.uid}')
f'Failed to gracefully terminate {proc}\n'
f"Attempting to hard kill {proc}")
finally:
log.runtime(f"Attempting to kill {proc}")
# XXX: do this **after** cancellation/tearfown
# to avoid killing the process too early
# since trio does this internally on ``__aexit__()``
await do_hard_kill(proc)
with trio.CancelScope(shield=True):
# XXX: do this **after**
# cancellation/tearfown to avoid killing the
# process too early since trio does this
# internally on ``__aexit__()``
await do_hard_kill(proc, hard_kill_after)
async def new_proc(
name: str,
actor_nursery: 'ActorNursery', # type: ignore # noqa
subactor: Actor,
errors: Dict[Tuple[str, str], Exception],
# passed through to actor main
bind_addr: Tuple[str, int],
parent_addr: Tuple[str, int],
_runtime_vars: Dict[str, Any], # serialized and sent to _child
*,
graceful_kill_timeout: int = 3,
infect_asyncio: bool = False,
task_status: TaskStatus[Portal] = trio.TASK_STATUS_IGNORED
) -> None:
"""Create a new ``multiprocessing.Process`` using the
spawn method as configured using ``try_set_start_method()``.
"""
cancel_scope = None
) -> None:
"""
Create a new ``multiprocessing.Process`` using the
spawn method as configured using ``try_set_start_method()``.
"""
# mark the new actor with the global spawn method
subactor._spawn_method = _spawn_method
uid = subactor.uid
if _spawn_method == 'trio':
async with trio.open_nursery() as nursery:
async with spawn_subactor(
subactor,
parent_addr,
infect_asyncio=infect_asyncio
) as proc:
log.runtime(f"Started {proc}")
# wait for actor to spawn and connect back to us
# channel should have handshake completed by the
# local actor by the time we get a ref to it
event, chan = await actor_nursery._actor.wait_for_peer(
spawn_cmd = [
sys.executable,
"-m",
# Hardcode this (instead of using ``_child.__name__`` to avoid a
# double import warning: https://stackoverflow.com/a/45070583
"tractor._child",
# We provide the child's unique identifier on this exec/spawn
# line for debugging purposes when viewing the process tree from
# the OS; it otherwise can be passed via the parent channel if
# we prefer in the future (for privacy).
"--uid",
str(uid),
# Address the child must connect to on startup
"--parent_addr",
str(parent_addr)
]
if subactor.loglevel:
spawn_cmd += [
"--loglevel",
subactor.loglevel
]
# Tell child to run in guest mode on top of ``asyncio`` loop
if infect_asyncio:
spawn_cmd.append("--asyncio")
proc = await trio.open_process(spawn_cmd)
log.info(f"Started {proc}")
portal: Optional[Portal] = None
try:
# wait for actor to spawn and connect back to us
# channel should have handshake completed by the
# local actor by the time we get a ref to it
event, chan = await actor_nursery._actor.wait_for_peer(
subactor.uid)
except trio.Cancelled:
# reap un-contacted process which are started
# but never setup a connection to parent.
log.warning(f'Spawning aborted due to cancel {proc}')
with trio.CancelScope(shield=True):
await do_hard_kill(proc, 0.1)
# this should break here
raise
actor_nursery_cancel_called = None
portal = Portal(chan)
actor_nursery._children[subactor.uid] = (
subactor, proc, portal)
# track child in current nursery
curr_actor = current_actor()
curr_actor._actoruid2nursery[subactor.uid] = actor_nursery
try:
# send additional init params
await chan.send({
"_parent_main_data": subactor._parent_main_data,
"enable_modules": subactor.enable_modules,
"_arb_addr": subactor._arb_addr,
"bind_host": bind_addr[0],
"bind_port": bind_addr[1],
"_runtime_vars": _runtime_vars,
})
# resume caller at next checkpoint now that child is up
task_status.started(portal)
# wait on actor nursery to complete
# with trio.CancelScope(shield=True):
# this either completes or is cancelled
# and should only arrive once the actor nursery
# has errored or exitted.
await actor_nursery._join_procs.wait()
except (
trio.Cancelled,
# KeyboardInterrupt,
# required to collect errors from multiple subactors
trio.MultiError,
# RuntimeError,
) as cerr:
actor_nursery_cancel_called = cerr
if actor_nursery.cancelled:
log.cancel(f'{uid}: nursery cancelled before exit')
else:
log.error(f'Child {uid} was cancelled before nursery exit?')
# we were specifically cancelled by our parent nursery
with trio.CancelScope(shield=True):
if portal.channel.connected():
log.cancel(f'Sending cancel IPC-msg to {uid}')
# try to cancel the actor @ IPC level
await portal.cancel_actor()
finally:
# 2 cases:
# - actor nursery was cancelled in which case
# we want to try a soft reap of the actor via
# ipc cancellation and then failing that do a hard
# reap.
# - this is normal termination and we must wait indefinitely
# for ria and daemon actors
reaping_cancelled: bool = False
ria = portal in actor_nursery._cancel_after_result_on_exit
# this is the soft reap sequence. we can
# either collect results:
# - ria actors get them them via ``Portal.result()``
# - we wait forever on daemon actors until they're
# cancelled by user code via ``Portal.cancel_actor()``
# or ``ActorNursery.cancel(). in the latter case
# we have to expect another cancel here since
# the task spawning nurseries will both be cacelled
# by ``ActorNursery.cancel()``.
# OR, we're cancelled while collecting results, which
# case we need to try another soft cancel and reap attempt.
try:
log.cancel(f'Starting soft actor reap for {uid}')
cancel_scope = None
async with trio.open_nursery() as nursery:
if ria:
# collect any expected ``.run_in_actor()`` results
cancel_scope = await nursery.start(
result_from_portal,
portal,
subactor,
errors,
True, # cancel_on_result
)
# soft & cancellable
await reap_proc(proc)
# if proc terminates before portal result
if cancel_scope:
cancel_scope.cancel()
except (
trio.Cancelled,
# is this required to collect errors from multiple subactors?
trio.MultiError,
) as rerr:
# nursery was closed but was cancelled during normal
# reaping.
reaping_cancelled = rerr
if actor_nursery.cancelled:
log.cancel(f'Nursery cancelled during soft reap for {uid}')
# hard reap sequence
if proc.poll() is None:
log.cancel('Attempting hard reap for {uid}')
# hard reap sequence
await maybe_wait_for_debugger()
with trio.CancelScope(shield=True):
if portal.channel.connected():
# cancel the process @ the IPC level
await portal.cancel_actor()
# TODO: do we need to try the ria portals
# again?
# await result_from_portal(
# portal,
# subactor,
# errors
# )
# hard zombie lord reap, with timeout
await reap_proc(
proc,
terminate_after=2,
)
finally:
# 2 cases:
# - the actor terminated gracefully
# - we're cancelled and likely need to re-raise
while proc.poll() is None:
log.critical("ZOMBIE LORD HAS ARRIVED for your {proc}")
with trio.CancelScope(shield=True):
await reap_proc(
proc,
terminate_after=0.1,
)
log.info(f"Joined {proc}")
# pop child entry to indicate we no longer managing this
# subactor
subactor, proc, portal = actor_nursery._children.pop(
subactor.uid)
portal = Portal(chan)
actor_nursery._children[subactor.uid] = (
subactor, proc, portal)
if not actor_nursery._children:
actor_nursery._all_children_reaped.set()
# send additional init params
await chan.send({
"_parent_main_data": subactor._parent_main_data,
"enable_modules": subactor.enable_modules,
"_arb_addr": subactor._arb_addr,
"bind_host": bind_addr[0],
"bind_port": bind_addr[1],
"_runtime_vars": _runtime_vars,
})
if actor_nursery_cancel_called:
raise actor_nursery_cancel_called
# track subactor in current nursery
curr_actor = current_actor()
curr_actor._actoruid2nursery[subactor.uid] = actor_nursery
if reaping_cancelled:
raise reaping_cancelled
# resume caller at next checkpoint now that child is up
task_status.started(portal)
# wait for ActorNursery.wait() to be called
with trio.CancelScope(shield=True):
await actor_nursery._join_procs.wait()
if portal in actor_nursery._cancel_after_result_on_exit:
cancel_scope = await nursery.start(
cancel_on_completion,
portal,
subactor,
errors
)
# Wait for proc termination but **dont' yet** call
# ``trio.Process.__aexit__()`` (it tears down stdio
# which will kill any waiting remote pdb trace).
# TODO: No idea how we can enforce zombie
# reaping more stringently without the shield
# we used to have below...
# with trio.CancelScope(shield=True):
# async with proc:
# Always "hard" join sub procs since no actor zombies
# are allowed!
# this is a "light" (cancellable) join, the hard join is
# in the enclosing scope (see above).
await proc.wait()
log.debug(f"Joined {proc}")
# pop child entry to indicate we no longer managing this subactor
subactor, proc, portal = actor_nursery._children.pop(subactor.uid)
# cancel result waiter that may have been spawned in
# tandem if not done already
if cancel_scope:
log.warning(
"Cancelling existing result waiter task for "
f"{subactor.uid}")
cancel_scope.cancel()
else:
# `multiprocessing`
# async with trio.open_nursery() as nursery:
@ -430,6 +606,9 @@ async def mp_new_proc(
# while user code is still doing it's thing. Only after the
# nursery block closes do we allow subactor results to be
# awaited and reported upwards to the supervisor.
# no shield is required here (vs. above on the trio backend)
# since debug mode is not supported on mp.
await actor_nursery._join_procs.wait()
finally:
@ -446,7 +625,7 @@ async def mp_new_proc(
# async with trio.open_nursery() as n:
# n.cancel_scope.shield = True
cancel_scope = await nursery.start(
cancel_on_completion,
result_from_portal,
portal,
subactor,
errors