jc211
/
tractor
forked from goodboy/tractor
1
0
Fork 0
Code Pull Requests Activity

Handle cancelled-before-proc-created spawn case 

It's definitely possible to have a nursery spawn task be cancelled
before a `trio.Process` handle is ever returned; we now handle this
case as a cancelled-during-spawn scenario. Zombie collection logic
also is bypassed in this case.
acked_backup
Tyler Goodlet 2021-11-05 10:43:14 -04:00
parent d05885d650
commit bf6958cdbe
1 changed files with 25 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

45
tractor/_spawn.py
View File

@ -230,17 +230,19 @@ async def new_proc(
] ]
cancelled_during_spawn: bool = False cancelled_during_spawn: bool = False
proc: Optional[trio.Process] = None
try: try:
proc = await trio.open_process(spawn_cmd)
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
try: try:
proc = await trio.open_process(spawn_cmd)
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( event, chan = await actor_nursery._actor.wait_for_peer(
subactor.uid) subactor.uid)
except trio.Cancelled: except trio.Cancelled:
cancelled_during_spawn = True cancelled_during_spawn = True
# we may cancel before the child connects back in which # we may cancel before the child connects back in which
@ -320,23 +322,26 @@ async def new_proc(
# killing the process too early. # killing the process too early.
log.cancel(f'Hard reap sequence starting for {uid}') log.cancel(f'Hard reap sequence starting for {uid}')
with trio.CancelScope(shield=True): if proc:
with trio.CancelScope(shield=True):
# don't clobber an ongoing pdb # don't clobber an ongoing pdb
if cancelled_during_spawn: if cancelled_during_spawn:
# Try again to avoid TTY clobbering. # Try again to avoid TTY clobbering.
async with acquire_debug_lock(uid): async with acquire_debug_lock(uid):
with trio.move_on_after(0.5): with trio.move_on_after(0.5):
await proc.wait() await proc.wait()
if is_root_process(): if is_root_process():
await maybe_wait_for_debugger() await maybe_wait_for_debugger()
if proc.poll() is None: if proc.poll() is None:
log.cancel(f"Attempting to hard kill {proc}") log.cancel(f"Attempting to hard kill {proc}")
await do_hard_kill(proc) await do_hard_kill(proc)
log.debug(f"Joined {proc}") log.debug(f"Joined {proc}")
else:
log.warning(f'Nursery cancelled before sub-proc started')
if not cancelled_during_spawn: if not cancelled_during_spawn:
# pop child entry to indicate we no longer managing this # pop child entry to indicate we no longer managing this