We Keep Coding

python thread not exiting with atexit

Here is my script. When I run it in a shell it just hangs indefinitely whereas I would expect it to terminate cleanly.
import logging
from logging import StreamHandler
import pymsteams
import queue
import threading
import atexit
class TeamsHandler(StreamHandler):
def __init__(self, channel_url):
super().__init__()
self.channel_url = channel_url
self.queue = queue.Queue()
self.thread = threading.Thread(target=self._worker)
self.thread.start()
atexit.register(self.queue.put, None)
def _worker(self):
while True:
record = self.queue.get()
if record is None:
break
msg = self.format(record)
print(msg)
def emit(self, record):
# enqueue the record to log and return control to the caller
self.queue.put(record)
if __name__ == "__main__":
my_logger = logging.getLogger('TestLogging')
my_logger.setLevel(logging.DEBUG)
console_handler = logging.StreamHandler()
console_handler.setLevel(logging.DEBUG)
my_logger.addHandler(console_handler)
CHANNEL_ID = "not_used_anyway"
teamshandler = TeamsHandler(CHANNEL_ID)
teamshandler.setFormatter(logging.Formatter('%(levelname)s %(message)s'))
teamshandler.setLevel(logging.DEBUG)
my_logger.addHandler(teamshandler)
for i in range(1, 2):
my_logger.error(f"this is an error [{i}]")
my_logger.info(f"this is an info [{i}]")
The None record that should be sent by atexit (line 28) never arrives so the thread stays open forever.
How to make sure that the program exits cleanly by modifying the TeamsHandler only ?

I got something working, have a look:
import queue
import threading
class Worker:
def __init__(self):
self.queue = queue.Queue()
threading.Thread(target=self._worker).start()
def _worker(self):
print("starting thread")
while True:
record = self.queue.get()
if record is None:
print("exiting")
break
print(f"Got message: {record}")
def emit(self, record):
self.queue.put(record)
class Wrapper:
def __init__(self):
self._worker = Worker()
def __del__(self):
print("Wrapper is being deleted")
self._worker.emit(None)
def emit(self, record):
self._worker.emit(record)
def main():
worker = Wrapper()
worker.emit("foo")
worker.emit("bar")
print("main exits")
if __name__ == "__main__":
main()
The point here is that when main exits, worker (which is an instance of Wrapper) goes out of scope, and its __del__ method is called, and it sends stop message to a real worker object.
The results of running this code ("Got message" lines can be in different places, of course):
starting thread
main exits
Wrapper is being deleted
Got message: foo
Got message: bar
exiting

As pointed out by avysk, the problem is likely that atexit handlers fire too late, after the waiting for the non-daemon threads is already (supposed to be) done, which leads to deadlock.
If I were you, I'd just add a call like TeamsHandler.finish() at the end of if __name__ == '__main__' block, and modify TeamsHandler along these lines (untested):
_queues = []
class TeamsHandler(StreamHandler):
def __init__(self, channel_url):
super().__init__()
self.channel_url = channel_url
self.queue = queue.Queue()
self.thread = threading.Thread(target=self._worker)
self.thread.start()
_queues.append(self.queue)
def _worker(self):
while True:
record = self.queue.get()
if record is None:
break
msg = self.format(record)
print(msg)
def emit(self, record):
# enqueue the record to log and return control to the caller
self.queue.put(record)
#staticmethod
def finish(self):
for q in _queues:
q.put(None)
del _queues[:]

Integrating multiprocessing.Process with concurrent.future._base.Future

I have a requirement of creating child processes, receive results using Future and then kill some of them when required.
For this I have subclassed multiprocessing.Process class and return a Future object from the start() method.
The problem is that I am not able to receive the result in the cb() function as it never gets called.
Please help/suggest if this can be done in some other way or something I am missing in my current implementation?
Following is my current approach
from multiprocessing import Process, Queue
from concurrent.futures import _base
import threading
from time import sleep
def foo(x,q):
print('result {}'.format(x*x))
result = x*x
sleep(5)
q.put(result)
class MyProcess(Process):
def __init__(self, target, args):
super().__init__()
self.target = target
self.args = args
self.f = _base.Future()
def run(self):
q = Queue()
worker_thread = threading.Thread(target=self.target, args=(self.args+ (q,)))
worker_thread.start()
r = q.get(block=True)
print('setting result {}'.format(r))
self.f.set_result(result=r)
print('done setting result')
def start(self):
f = _base.Future()
run_thread = threading.Thread(target=self.run)
run_thread.start()
return f
def cb(future):
print('received result in callback {}'.format(future))
def main():
p1 = MyProcess(target=foo, args=(2,))
f = p1.start()
f.add_done_callback(fn=cb)
sleep(10)
if __name__ == '__main__':
main()
print('Main thread dying')

In your start method you create a new Future which you then return. This is a different future then the one you set the result on, this future is just not used at all. Try:
def start(self):
run_thread = threading.Thread(target=self.run)
run_thread.start()
return self.f
However there are more problems with your code. You override the start method of the process, replacing it with execution on a worker thread, therefore actually bypassing multiprocessing. Also you shouldn't import the _base module, that is an implementation detail as seen from the leading underscore. You should import concurrent.futures.Future (it's the same class, but through public API).
This really uses multiprocessing:
from multiprocessing import Process, Queue
from concurrent.futures import Future
import threading
from time import sleep
def foo(x,q):
print('result {}'.format(x*x))
result = x*x
sleep(5)
q.put(result)
class MyProcess(Process):
def __init__(self, target, args):
super().__init__()
self.target = target
self.args = args
self.f = Future()
def run(self):
q = Queue()
worker_thread = threading.Thread(target=self.target, args=(self.args+ (q,)))
worker_thread.start()
r = q.get(block=True)
print('setting result {}'.format(r))
self.f.set_result(result=r)
print('done setting result')
def cb(future):
print('received result in callback {}: {}'.format(future, future.result()))
def main():
p1 = MyProcess(target=foo, args=(2,))
p1.f.add_done_callback(fn=cb)
p1.start()
p1.join()
sleep(10)
if __name__ == '__main__':
main()
print('Main thread dying')
And you're already in a new process now, spawning a worker thread to execute your target function shouldn't really be necessary, you could just execute your target function directly instead. Should the target function raise an Exception you wouldn't know about it, your callback will only be called on success. So if you fix that, then you're left with:
from multiprocessing import Process
from concurrent.futures import Future
import threading
from time import sleep
def foo(x):
print('result {}'.format(x*x))
result = x*x
sleep(5)
return result
class MyProcess(Process):
def __init__(self, target, args):
super().__init__()
self.target = target
self.args = args
self.f = Future()
def run(self):
try:
r = self.target(*self.args)
print('setting result {}'.format(r))
self.f.set_result(result=r)
print('done setting result')
except Exception as ex:
self.f.set_exception(ex)
def cb(future):
print('received result in callback {}: {}'.format(future, future.result()))
def main():
p1 = MyProcess(target=foo, args=(2,))
p1.f.add_done_callback(fn=cb)
p1.start()
p1.join()
sleep(10)
if __name__ == '__main__':
main()
print('Main thread dying')
This is basically what a ProcessPoolExecutor does.

Python send variables to thread

I'm trying to create my own threading class in Python2.7. I want it to be able to stop that thread with my own class function. Currently I have something like this:
class loop(threading.Thread):
def __init__(self, myvar):
super(loop, self).__init__()
self.terminate = False
self.myvar = myvar
def run(self):
while not self.terminate:
do.smthng.useful(self.myvar)
def change(self, newvar):
self.myvar = newvar #Doesnt work, in run() my old var is still being used
def stoploop(self):
self.terminate = True #Also not working
l = loop(1)
l.start()
time.sleep(1)
l.change(2) #thread still using "1"
time.sleep(1)
l.stoploop() #doesnt stop
I've read some posts here about this, but it wasnt what I needed.
Any help would be appreciated.
Thank you.
EDIT:
As some of the commenters already stated, this part of code looks like to be really working! Problem is in another place of my project. I've found it, but can't solve it. Maybe some of you could help.
So, my project uses Apache Thrift library and the server is in python.
Server.py:
loo = loop(0)
handler = ServHandler(loo)
processor = serv.Processor(handler)
transport = TSocket.TServerSocket('0.0.0.0', port=9090)
tfactory = TTransport.TBufferedTransportFactory()
pfactory = TBinaryProtocol.TBinaryProtocolFactory()
server = TProcessPoolServer.TProcessPoolServer(processor, transport, tfactory, pfactory)
print 'Starting the server...'
server.serve()
ServHandler.py:
class ServHandler:
def __init__(self, loo):
self.loo = loo
def terminate(self): #Function that can be called remotely
self.loo.stoploop() #Doesn't work
In above case thread isn't terminated and I don't why. There's no error, object exists, but it sets self.terminate value somewhere else. The object id seems to be the same as well as memory address, but it just looks like object is different although loop init function is called only once...
Below is the example, when the loop is terminated successfully.
ServHandler.py:
class ServHandler:
def __init__(self, loo):
self.loo = None
def terminate(self): #Function that can be called remotely
self.loo.stoploop() #Does work!!!!!!
def create(self):
self.loo = loop(0) #Function that can be called remotely
When I create loop object remotely, I can terminate it remotely. But it doesn't fit me. There should be a thread created before thrift server is served and multiple users have to be able to change vars/terminate/etc of that thread. How can I achieve this?
Thank you!

Not a answer per sae, but a useful debug code for the OP
from time import sleep
from threading import Thread
class loop(Thread):
def __init__(self, myvar):
Thread.__init__(self)
self.terminate = False
self.myvar = myvar
def run(self):
while self.terminate is False:
print('Run says myvar is:',self.myvar)
sleep(0.5)
def change(self, newvar):
self.myvar = newvar
def stoploop(self):
self.terminate = True
l = loop(1)
l.start()
sleep(1)
l.change(2)
sleep(1)
l.stoploop()
print('Final product:',l.myvar)
sleep(2)
print('Is the thread alive:',l.isAlive())
Tried your code with some debugging prints, and it's working?
Following code produced:
[torxed#archie ~]$ python test.py
Run says myvar is: 1
Run says myvar is: 1
Run says myvar is: 2 <-- Proves that change() does change `myvar`
Run says myvar is: 2
Final product: 2 <-- Also the global scope knows about the change
Is the thread alive: False <-- And the thread got terminated as intended
However, these are not bulletproof ideas when fetching data or dealing with thread-returns for a number of reasons (even tho i use this method myself from time to time), you should consider using thread.join which should be used in combination with l.toplooop() like so:
l = loop(1)
l.start()
l.change(2)
l.stoploop()
ret = l.join()
Also when updating data you should aquire locks on your data so collisions don't occur, have a look at semaphore objects.

Is it what you need?
import threading
import time
class Worker(threading.Thread):
def __init__(self):
threading.Thread.__init__(self)
self.state = threading.Condition()
self.variable = 10
self.paused = False
def run(self):
while True:
with self.state:
if self.paused:
self.state.wait()
self.do_stuff()
def do_stuff(self):
time.sleep(.1)
print self.variable
def resume(self):
with self.state:
self.paused = False
self.state.notify()
def pause(self):
with self.state:
self.paused = True
loop = Worker()
loop.start()
time.sleep(1)
loop.pause()
loop.variable = 11
print 'CHANGED!'
loop.resume()
time.sleep(1)

Python multiprocessing RemoteManager under a multiprocessing.Process

I'm trying to start a data queue server under a managing process (so that it can later be turned into a service), and while the data queue server function works fine in the main process, it does not work in a process created using multiprocessing.Process.
The dataQueueServer and dataQueueClient code is based on the code from the multiprocessing module documentation here.
When run on its own, dataQueueServer works well. However, when run using a multiprocessing.Process's start() in mpquueue, it doesn't work (when tested with the client). I am using the dataQueueClient without changes to test both cases.
The code does reach the serve_forever in both cases, so I think the server is working, but something is blocking it from communicating back to the client in the mpqueue case.
I have placed the loop that runs the serve_forever() part under a thread, so that it can be stoppable.
Here is the code:
mpqueue # this is the "manager" process trying to spawn the server in a child process
import time
import multiprocessing
import threading
import dataQueueServer
class Printer():
def __init__(self):
self.lock = threading.Lock()
def tsprint(self, text):
with self.lock:
print text
class QueueServer(multiprocessing.Process):
def __init__(self, name = '', printer = None):
multiprocessing.Process.__init__(self)
self.name = name
self.printer = printer
self.ml = dataQueueServer.MainLoop(name = 'ml', printer = self.printer)
def run(self):
self.printer.tsprint(self.ml)
self.ml.start()
def stop(self):
self.ml.stop()
if __name__ == '__main__':
printer = Printer()
qs = QueueServer(name = 'QueueServer', printer = printer)
printer.tsprint(qs)
printer.tsprint('starting')
qs.start()
printer.tsprint('started.')
printer.tsprint('Press Ctrl-C to quit')
try:
while True:
time.sleep(60)
except KeyboardInterrupt:
printer.tsprint('\nTrying to exit cleanly...')
qs.stop()
printer.tsprint('stopped')
dataQueueServer
import time
import threading
from multiprocessing.managers import BaseManager
from multiprocessing import Queue
HOST = ''
PORT = 50010
AUTHKEY = 'authkey'
## Define some helper functions for use by the main process loop
class Printer():
def __init__(self):
self.lock = threading.Lock()
def tsprint(self, text):
with self.lock:
print text
class QueueManager(BaseManager):
pass
class MainLoop(threading.Thread):
"""A thread based loop manager, allowing termination signals to be sent
to the thread"""
def __init__(self, name = '', printer = None):
threading.Thread.__init__(self)
self._stopEvent = threading.Event()
self.daemon = True
self.name = name
if printer is None:
self.printer = Printer()
else:
self.printer = printer
## create the queue
self.queue = Queue()
## Add a function to the handler to return the queue to clients
self.QM = QueueManager
self.QM.register('get_queue', callable=lambda:self.queue)
self.queue_manager = self.QM(address=(HOST, PORT), authkey=AUTHKEY)
self.queue_server = self.queue_manager.get_server()
def __del__(self):
self.printer.tsprint( 'closing...')
def run(self):
self.printer.tsprint( '{}: started serving'.format(self.name))
self.queue_server.serve_forever()
def stop(self):
self.printer.tsprint ('{}: stopping'.format(self.name))
self._stopEvent.set()
def stopped(self):
return self._stopEvent.isSet()
def start():
printer = Printer()
ml = MainLoop(name = 'ml', printer = printer)
ml.start()
return ml
def stop(ml):
ml.stop()
if __name__ == '__main__':
ml = start()
raw_input("\nhit return to stop")
stop(ml)
And a client:
dataQueueClient
import datetime
from multiprocessing.managers import BaseManager
n = 0
N = 10**n
HOST = ''
PORT = 50010
AUTHKEY = 'authkey'
def now():
return datetime.datetime.now()
def gen(n, func, *args, **kwargs):
k = 0
while k < n:
yield func(*args, **kwargs)
k += 1
class QueueManager(BaseManager):
pass
QueueManager.register('get_queue')
m = QueueManager(address=(HOST, PORT), authkey=AUTHKEY)
m.connect()
queue = m.get_queue()
def load(msg, q):
return q.put(msg)
def get(q):
return q.get()
lgen = gen(N, load, msg = 'hello', q = queue)
t0 = now()
while True:
try:
lgen.next()
except StopIteration:
break
t1 = now()
print 'loaded %d items in ' % N, t1-t0
t0 = now()
while queue.qsize() > 0:
queue.get()
t1 = now()
print 'got %d items in ' % N, t1-t0

So it seems like the solution is simple enough: Don't use serve_forever(), and use manager.start() instead.
According to Eli Bendersky, the BaseManager (and it's extended version SyncManager) already spawns the server in a new process (and looking at the multiprocessing.managers code confirms this). The problem I have been experiencing stems from the form used in the example, in which the server is started under the main process.
I still don't understand why the current example doesn't work when run under a child process, but that's no longer an issue.
Here's the working (and much simplified from OP) code to manage multiple queue servers:
Server:
from multiprocessing import Queue
from multiprocessing.managers import SyncManager
HOST = ''
PORT0 = 5011
PORT1 = 5012
PORT2 = 5013
AUTHKEY = 'authkey'
name0 = 'qm0'
name1 = 'qm1'
name2 = 'qm2'
description = 'Queue Server'
def CreateQueueServer(HOST, PORT, AUTHKEY, name = None, description = None):
name = name
description = description
q = Queue()
class QueueManager(SyncManager):
pass
QueueManager.register('get_queue', callable = lambda: q)
QueueManager.register('get_name', callable = name)
QueueManager.register('get_description', callable = description)
manager = QueueManager(address = (HOST, PORT), authkey = AUTHKEY)
manager.start() # This actually starts the server
return manager
# Start three queue servers
qm0 = CreateQueueServer(HOST, PORT0, AUTHKEY, name0, description)
qm1 = CreateQueueServer(HOST, PORT1, AUTHKEY, name1, description)
qm2 = CreateQueueServer(HOST, PORT2, AUTHKEY, name2, description)
raw_input("return to end")
Client:
from multiprocessing.managers import SyncManager
HOST = ''
PORT0 = 5011
PORT1 = 5012
PORT2 = 5013
AUTHKEY = 'authkey'
def QueueServerClient(HOST, PORT, AUTHKEY):
class QueueManager(SyncManager):
pass
QueueManager.register('get_queue')
QueueManager.register('get_name')
QueueManager.register('get_description')
manager = QueueManager(address = (HOST, PORT), authkey = AUTHKEY)
manager.connect() # This starts the connected client
return manager
# create three connected managers
qc0 = QueueServerClient(HOST, PORT0, AUTHKEY)
qc1 = QueueServerClient(HOST, PORT1, AUTHKEY)
qc2 = QueueServerClient(HOST, PORT2, AUTHKEY)
# Get the queue objects from the clients
q0 = qc0.get_queue()
q1 = qc1.get_queue()
q2 = qc2.get_queue()
# put stuff in the queues
q0.put('some stuff')
q1.put('other stuff')
q2.put({1:123, 2:'abc'})
# check their sizes
print 'q0 size', q0.qsize()
print 'q1 size', q1.qsize()
print 'q2 size', q2.qsize()
# pull some stuff and print it
print q0.get()
print q1.get()
print q2.get()
Adding an additional server to share a dictionary with the information of the running queue servers so that consumers can easily tell what's available where is easy enough using that model. One thing to note, though, is that the shared dictionary requires slightly different syntax than a normal dictionary: dictionary[0] = something will not work. You need to use dictionary.update([(key, value), (otherkey, othervalue)]) and dictionary.get(key) syntax, which propagates across to all other clients connected to this dictionary..

" RuntimeError: thread.__init__() not called" when subclassing threading.Thread

I need to run as many threads of class Observer as there are elements in list dirlist.
When I run it python console it works all right.
class Observer(Thread):
def run(self):
naptime = random.randint(1,10)
print(self.name + ' starting, running for %ss.' % naptime)
time.sleep(naptime)
print(self.name + ' done')
observers = {}
for d in dirlist:
observers[d] = Observer()
observers[d].start()
But when I try to do it from a Master thread which is supposed to spawn the Observer threads I get errors.
class Master(Thread):
def __init__(self, dirlist):
self.dirlist = dirlist
def run(self):
observers = {}
for d in dirlist:
observers[d] = Observer()
observers[d].start()
while True:
time.sleep(3600)
master_thread = Master(dirlist)
master_thread.start()
The call to Master.start results in:
RuntimeError: thread.__init__() not called
This looks strange to me.
I am unable to understand whats the difference between both cases.
Can anybody figure out a solution to my problem ?
Somehow following doesn't produce an error, and I don't understand why.
class Master(Thread):
def set(self, dirlist):
self.dirlist = dirlist
def run(self):
observers = {}
for d in dirlist:
observers[d] = Observer()
observers[d].start()
while True:
time.sleep(3600)
master_thread = Master()
master_thread.set(dirlist)
master_thread.start()

>>> master_thread.start()
RuntimeError: thread.__init__() not called
Make sure to call Thread.__init__() in your Master.__init__:
class Master(Thread):
def __init__(self, dirlist):
super(Master, self).__init__()
self.dirlist = dirlist

Well i know it's late to answer but, what the hell, i am a newbie in python but this same thing was happening to me, so i went back to read the python tutorial and it has some differences with what we both we're trying, hope it helps.
instead of this
import threading
class Master(Thread):
def set(self, dirlist):
self.dirlist = dirlist
def run(self):
observers = {}
for d in dirlist:
...
class according to the python tutorial:
class Master(threading.Thread):
this line is missing:
threading.Thread.__init__(self)
so it will end up being:
import threading
class Master(threading.Thread):
def __init__(self, dirlist):
threading.Thread.__init__(self)
self.dirlist = dirlist
def run(self):
observers = {}
for d in dirlist:
...
and that should work, at least work for me.
I hope it was helpfull.
And your second try using the set method works, because you are not overriding the
__init__ method
from Thread therefore the original init method from the parent class is used it runs as it's supposed.

Error is clear, you should call thread.__init__():
def __init__(self, dirlist):
super(Master, self).__init__()
self.dirlist = dirlist