Develop Reference

Thread does not end cleanly due to socket recv callback

I am trying to close the socket and terminate the thread when shutdown handler callback is executed (I send SIGINT signal by CTRL+C)
main
from example import Example
if __name__ == '__main__':
exmpl = Example()
success = exmpl.connect('', 2000)
if success:
rospy.on_shutdown(exmpl.shutdown_handler)
rospy.spin()
imported class
import socket
import threading
class Example(object):
def connect(self, host='', port=2000):
try:
self.sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
self.sock.settimeout(10)
self.sock.bind((host, port))
self.sock.listen(1)
self.conn, self.addr = self.sock.accept()
self.sock.settimeout(None)
self.t = threading.Thread(target=self.recv_msg)
self.t.start()
return True
except Exception as e:
return False
except Exception as timeout:
return False
def recv_msg(self):
while True:
recv_msg = self.conn.recv(1024)
print(recv_msg)
def shutdown_handler(self):
try:
self.sock.close()
self.t.join()
except Exception as e:
print(e)
In this application recv_msg is always executing because data are coming all the time. So, when I send SIGINT, shutdown handler starts and only executes the statement sock.close() but not t.join() and thread never ends

chat-room server not closing threads and not exiting when SIGINT

I have the following server program in Python which simulates a chat-room. The code accepts connections from clients and for each of them it launches a new thread. This thread will wait for messages from this client. The messages can be L so that the server will respond with a list of connected clients, ip:port msg the server will send the message msg to the client ip:port.
On client side there will be 2 threads, one for receiving messages from the server, the other for sending.
import socket
from threading import Thread
#from SocketServer import ThreadingMixIn
import signal
import sys
import errno
EXIT = False
address = []
address2 = []
# handler per il comando Ctrl+C
def sig_handler(signum, frame):
if (signum == 2):
print("Called SIGINT")
EXIT = True
signal.signal(signal.SIGINT, sig_handler) # setto l'handler per i segnali
# Multithreaded Python server : TCP Server Socket Thread Pool
class ClientThread(Thread):
def __init__(self,conn,ip,port):
Thread.__init__(self)
self.conn = conn
self.ip = ip
self.port = port
print ("[+] New server socket thread started for " + ip + ":" + str(port))
def run(self):
while True:
data = self.conn.recv(1024)
print ("Server received data:", data)
if (data=='L'):
#print "QUI",address2
tosend = ""
for i in address2:
tosend = tosend + "ip:"+str(i[0]) + "port:"+str(i[1])+"\n"
self.conn.send(tosend)
#mandare elenco client connessi
else:
#manda ip:port msg
st = data.split(" ")
msg = st[1:]
msg = ' '.join(msg)
print ("MSG 2 SEND: ",msg)
ipport = st[0].split(":")
ip = ipport[0]
port = ipport[1]
flag = False
print ("Address2:",address2)
print ("ip:",ip)
print ("port:",port)
for i in address2:
print (i[0],ip,type(i[0]),type(ip),i[1],type(i[1]),port,type(port))
if str(i[0])==str(ip) and str(i[1])==str(port):
i[2].send(msg)
self.conn.send("msg inviato")
flag = True
break
if flag == False:
self.conn.send("client non esistente")
if __name__ == '__main__':
# Multithreaded Python server : TCP Server Socket Program Stub
TCP_IP = '127.0.0.1'
TCP_PORT = 2004
TCP_PORTB = 2005
BUFFER_SIZE = 1024 # Usually 1024, but we need quick response
tcpServer = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
tcpServer.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
tcpServer.bind((TCP_IP, TCP_PORT))
tcpServerB = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
tcpServerB.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
tcpServerB.bind((TCP_IP, TCP_PORTB))
threads = []
tcpServer.listen(4)
tcpServerB.listen(4)
while True:
print("Multithreaded Python server : Waiting for connections from TCP clients...")
try:
(conn, (ip,port)) = tcpServer.accept()
except socket.error as e: #(code, msg):
if e.errno != errno.EINTR:
raise
else:
break
address.append((ip,port,conn))
(conn2, (ip2,port2)) = tcpServerB.accept()
address2.append((ip2,port2,conn2))
newthread = ClientThread(conn,ip,port)
newthread.start()
threads.append(newthread)
if EXIT==True:
break
print ("SERVER EXIT")
for t in threads:
t.join()
The code has a signal handler for SIGINT to make the exit cleaner (closing connections, sending a message to the client (still to be implemented) and so on ). The handler writes a global flag EXIT to make the infinite loops terminate.
The code runs both in Python2 and Python3. However there are some problems with SIGINT signal generated by CTRL-C. When there is no client connected the program launched with Python2 exits correctly while the one in Python3 does not. Why this behavioural difference?
Considering only running the program in Python2, when a client connects and I press CTRL-C, the main while exits, like the signal is catched always by the main thread and this interrupts the blocking system call accept. However the other threads do not, I think because of the blocking underlying system call data = self.conn.recv(1024). In C I would block SIGINT signals for one thread and then call pthread_cancel from the other thread. How to exit from all threads when SIGINT is generated in Python?
The client program that for the moment works in Python2 only and suffers from the same problem is:
# Python TCP Client A
import socket
from threading import Thread
class ClientThread(Thread):
def __init__(self,conn):
Thread.__init__(self)
self.conn = conn
def run(self):
while True:
data = self.conn.recv(1024)
print "Ricevuto msg:",data
host = socket.gethostname()
print "host:",host
port = 2004
portB = 2005
BUFFER_SIZE = 2000
tcpClientA = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
tcpClientB = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
tcpClientA.connect(('127.0.0.1', port))
tcpClientB.connect(('127.0.0.1', portB))
newthread = ClientThread(tcpClientB)
newthread.start()
while(True):
msg = raw_input("Inserisci comando: ")
tcpClientA.send (msg)
data = tcpClientA.recv(BUFFER_SIZE)
print "data received:",data
tcpClientA.close()

As for the difference in behavior with accept() in Python 3, look at the full description in the docs. I think this is the key statement:
Changed in version 3.5: If the system call is interrupted and the signal handler does not raise an exception, the method now retries the system call instead of raising an InterruptedError exception (see PEP 475 for the rationale).
The other problem, stated in your penultimate sentence:
How to exit from all threads when SIGINT is generated in Python 2?
Take a look at the threading documentation:
A thread can be flagged as a “daemon thread”. The significance of this flag is that the entire Python program exits when only daemon threads are left. The initial value is inherited from the creating thread. The flag can be set through the daemon property.

Sending data from outside socket object in python

I made a Client socket object, which I instantiate and it keeps alive a connection with the server, which is working fine, but I'm wondering if there is a way to call the socket.send event from outside the instance. I was about to make a stack for the messages and check the stack in the while loop and if it's not empty then send the oldest data to the server, which would be just fine for me, but my problem is that the stack only updates after the while loop(I tried breaking out, then it updated).
So my question would be, is there a way to update the global stack simultaneously with the while loop running? Or is there any other way to call the socket.send event outside the object?
import socket
import sys
import select
import threading
SERVER_IP = '192.168.1.4'
PORT = 8686
TIMEOUT = 5
BUF_SIZE = 1024
MESSAGES = ['testdata1', 'testdata2']
class Client(threading.Thread):
def __init__(self, host=SERVER_IP, port=PORT):
threading.Thread.__init__(self)
self.sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
self.sock = socket.create_connection((host, port), 1)
self.sock.setblocking(0)
while 1:
try:
global MESSAGES
ready = select.select([self.sock], [], [], TIMEOUT*1000)
if ready[0]:
buf = self.sock.recv(BUF_SIZE)
print buf
#TODO:do stuff with buf
print 'messages left:'+str(len(MESSAGES))
if len(MESSAGES)>0:
self.sock.send(MESSAGES.pop())
except KeyboardInterrupt:
self.sock.close()
sys.exit(1)
except Exception, e:
print '\n[ERR] %s' % e
self.sock.close()
sys.exit(1)
def run(self):
pass
def sendData(self, data):
global MESSAGES
print 'appending data:%s' % data
MESSAGES.append(data)
def main():
client = Client()
client.start()
client.sendData("test1")
client.sendData("test2")
client.sendData("test3")
if __name__ == '__main__':
try:
main()
except KeyboardInterrupt:
sys.exit(1)

Client.__init__() does not return because it enters an infinite while loop. Hence control is never returned to the main thread, and the Client thread is not actually started.
Instead you should move the while loop into the run() method. Then the __init__() method will return control to the main thread, which can then start the thread, and request that the client send messages via sendData().
class Client(threading.Thread):
def __init__(self, host=SERVER_IP, port=PORT):
threading.Thread.__init__(self)
self.sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
self.sock = socket.create_connection((host, port), 1)
self.sock.setblocking(0)
def run(self):
while 1:
try:
global MESSAGES
ready = select.select([self.sock], [], [], TIMEOUT*1000)
if ready[0]:
buf = self.sock.recv(BUF_SIZE)
print buf
#TODO:do stuff with buf
print 'messages left:'+str(len(MESSAGES))
if len(MESSAGES)>0:
self.sock.send(MESSAGES.pop())
except KeyboardInterrupt:
self.sock.close()
sys.exit(1)
except Exception, e:
print '\n[ERR] %s' % e
self.sock.close()
sys.exit(1)
def sendData(self, data):
global MESSAGES
print 'appending data:%s' % data
MESSAGES.append(data)
Instead of using the global MESSAGES list you should probably create a Queue for communicating between the main thread and the worker thread(s), particularly if more than one worker thread is running. Something like this (untested!):
import Queue
class Client(threading.Thread):
def __init__(self, msg_queue, host=SERVER_IP, port=PORT):
threading.Thread.__init__(self)
self.msg_queue = msg_queue
self.sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
self.sock = socket.create_connection((host, port), 1)
self.sock.setblocking(0)
def run(self):
while 1:
try:
ready = select.select([self.sock], [], [], TIMEOUT*1000)
if ready[0]:
buf = self.sock.recv(BUF_SIZE)
print buf
#TODO:do stuff with buf
print 'messages left:'+ str(self.msg_queue.qsize())
try:
msg = self.msg_queue.get_nowait()
self.sock.send(msg)
except Queue.Empty:
pass
except KeyboardInterrupt:
self.sock.close()
sys.exit(1)
except Exception, e:
print '\n[ERR] %s' % e
self.sock.close()
sys.exit(1)
def main():
# create a queue and pass it to the client
msg_queue = Queue.Queue()
client = Client(msg_queue)
client.start()
msg_queue.put("test1")
msg_queue.put("test2")
msg_queue.put("test3")

The thing should work if you move your loop from
__init__() into run()
method instead.
Your thread is not a thread this way, process blocks at client = Client(...).
Why do you mix select and threads? Is this really necessary? If you want asynchronous sending and receiving without threads use asyncore module.
Or remove select from your code. The socket.recv() will block until it receives data in blocking mode, but as this is a thread, I don't see anything wrong about that. If in nonblocking mode, recv() will just return None if there is no data to receive if I remember correctly. So you don't really need select. Just check if recv() returned None. If it does, sleep some time before trying again.
The way you did it troubles your OS twice. Once for reading a socket, and second time to get the status of a socket where timeout is used to simulate sleep() more than anything else. Then the loop checks again making select() system call right after timeout confirmed that there is nothing to do for that socket.

Terminating process from Timer thread

I would better start the question from the code.
from multiprocessing import Process, Event, Queue
from threading import Timer
from Queue import Empty
class MyServer(Process):
def __init__(self, port, queue):
Process.__init__(self)
self.port = port
self.queue = queue
self.sd = None
def run(self):
try:
self.start_serving()
except KeyboardInterrupt:
print("Shutting down..")
finally:
if self.sd is not None:
self.sd.close()
def start_serving(self):
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
self.sd = s
try:
s.bind(('', self.port))
s.listen(1)
while True:
conn, addr = s.accept()
while True:
# I dont want to bore you with excess code
# just recv data from clients
try:
msg = self.queue.get_nowait()
# here i start Timer with delay from message (I'll describe Message class below)
Timer(msg.delay, self.response_handler, args=(conn, msg)).start()
except Empty:
pass
conn.close()
finally:
s.close()
def response_handler(self, sd, msg):
# doesn't matter
# and now I want to terminate the MyServer instance
if msg.terminate:
# the problem is here. Lets call it 'problem line'
sys.exit()
msg is instance of Message class which is:
class Message(object):
def __init__(self, port, delay, values, terminate=False):
self.port = port
self.delay = delay
self.values = values
self.terminate = terminate
The logic is I get data from clients via TCP connection and check Queue for message. Messages are things to control the server. Sometimes I get a message like "wait 3 seconds and terminate the server".
What I have done so far.
Call self.terminate() at the problem line. I get
AttributeError: 'NoneType' object has no attribute 'terminate'
Raise an exception at the problem line. I assumed the exception was caught in run() function. I was
wrong
Call sys.exit(). It doesn't work too.
Perhaps my question can be shorter. How to terminate the process from its thread in Python?

Why don't you use multiprocessing.Event (you are already importing it) and exit the process gracefully if you get an terminate message.
To do this add this to __init__:
self.exit = Event()
And change the two while loops:
while True:
conn, addr = s.accept()
while True:
#...
to
while not self.exit.is_set():
conn, addr = s.accept()
while not self.exit.is_set()
#...
then in your response handler:
if msg.terminate:
self.exit.set()
this will allow the code to naturally exit the loops, ensuring that conn.close() is called.

How do I abort a socket.recvfrom() from another thread in python?

This looks like a duplicate of How do I abort a socket.recv() from another thread in Python, but it's not, since I want to abort recvfrom() in a thread, which is UDP, not TCP.
Can this be solved by poll() or select.select() ?

If you want to unblock a UDP read from another thread, send it a datagram!
Rgds,
Martin

A good way to handle this kind of asynchronous interruption is the old C pipe trick. You can create a pipe and use select/poll on both socket and pipe: Now when you want interrupt receiver you can just send a char to the pipe.
pros:
Can work both for UDP and TCP
Is protocol agnostic
cons:
select/poll on pipes are not available on Windows, in this case you should replace it by another UDP socket that use as notification pipe
Starting point
interruptable_socket.py
import os
import socket
import select
class InterruptableUdpSocketReceiver(object):
def __init__(self, host, port):
self._host = host
self._port = port
self._socket = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
self._socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
self._r_pipe, self._w_pipe = os.pipe()
self._interrupted = False
def bind(self):
self._socket.bind((self._host, self._port))
def recv(self, buffersize, flags=0):
if self._interrupted:
raise RuntimeError("Cannot be reused")
read, _w, errors = select.select([self._r_pipe, self._socket], [], [self._socket])
if self._socket in read:
return self._socket.recv(buffersize, flags)
return ""
def interrupt(self):
self._interrupted = True
os.write(self._w_pipe, "I".encode())
A test suite:
test_interruptable_socket.py
import socket
from threading import Timer
import time
from interruptable_socket import InterruptableUdpSocketReceiver
import unittest
class Sender(object):
def __init__(self, destination_host, destination_port):
self._socket = socket.socket(socket.AF_INET, socket.SOCK_DGRAM, socket.IPPROTO_UDP)
self._dest = (destination_host, destination_port)
def send(self, message):
self._socket.sendto(message, self._dest)
class Test(unittest.TestCase):
def create_receiver(self, host="127.0.0.1", port=3010):
receiver = InterruptableUdpSocketReceiver(host, port)
receiver.bind()
return receiver
def create_sender(self, host="127.0.0.1", port=3010):
return Sender(host, port)
def create_sender_receiver(self, host="127.0.0.1", port=3010):
return self.create_sender(host, port), self.create_receiver(host, port)
def test_create(self):
self.create_receiver()
def test_recv_async(self):
sender, receiver = self.create_sender_receiver()
start = time.time()
send_message = "TEST".encode('UTF-8')
Timer(0.1, sender.send, (send_message, )).start()
message = receiver.recv(128)
elapsed = time.time()-start
self.assertGreaterEqual(elapsed, 0.095)
self.assertLess(elapsed, 0.11)
self.assertEqual(message, send_message)
def test_interrupt_async(self):
receiver = self.create_receiver()
start = time.time()
Timer(0.1, receiver.interrupt).start()
message = receiver.recv(128)
elapsed = time.time()-start
self.assertGreaterEqual(elapsed, 0.095)
self.assertLess(elapsed, 0.11)
self.assertEqual(0, len(message))
def test_exception_after_interrupt(self):
sender, receiver = self.create_sender_receiver()
receiver.interrupt()
with self.assertRaises(RuntimeError):
receiver.recv(128)
if __name__ == '__main__':
unittest.main()
Evolution
Now this code is just a starting point. To make it more generic I see we should fix follow issues:
Interface: return empty message in interrupt case is not a good deal, is better to use an exception to handle it
Generalization: we should have just a function to call before socket.recv(), extend interrupt to others recv methods become very simple
Portability: to make simple port it to windows we should isolate the async notification in a object to choose the right implementation for our operating system
First of all we change test_interrupt_async() to check exception instead empty message:
from interruptable_socket import InterruptException
def test_interrupt_async(self):
receiver = self.create_receiver()
start = time.time()
with self.assertRaises(InterruptException):
Timer(0.1, receiver.interrupt).start()
receiver.recv(128)
elapsed = time.time()-start
self.assertGreaterEqual(elapsed, 0.095)
self.assertLess(elapsed, 0.11)
After this we can replace return '' by raise InterruptException and the tests pass again.
The ready to extend version can be :
interruptable_socket.py
import os
import socket
import select
class InterruptException(Exception):
pass
class InterruptableUdpSocketReceiver(object):
def __init__(self, host, port):
self._host = host
self._port = port
self._socket = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
self._socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
self._async_interrupt = AsycInterrupt(self._socket)
def bind(self):
self._socket.bind((self._host, self._port))
def recv(self, buffersize, flags=0):
self._async_interrupt.wait_for_receive()
return self._socket.recv(buffersize, flags)
def interrupt(self):
self._async_interrupt.interrupt()
class AsycInterrupt(object):
def __init__(self, descriptor):
self._read, self._write = os.pipe()
self._interrupted = False
self._descriptor = descriptor
def interrupt(self):
self._interrupted = True
self._notify()
def wait_for_receive(self):
if self._interrupted:
raise RuntimeError("Cannot be reused")
read, _w, errors = select.select([self._read, self._descriptor], [], [self._descriptor])
if self._descriptor not in read:
raise InterruptException
def _notify(self):
os.write(self._write, "I".encode())
Now wraps more recv function, implement a windows version or take care of socket timeouts become really simple.

The solution here is to forcibly close the socket. The problem is that the method for doing this is OS-specific and Python does not do a good job of abstracting the way to do it or the consequences. Basically, you need to do a shutdown() followed by a close() on the socket. On POSIX systems such as Linux, the shutdown is the key element in forcing recvfrom to stop (a call to close() alone won't do it). On Windows, shutdown() does not affect the recvfrom and the close() is the key element. This is exactly the behavior that you would see if you were implementing this code in C and using either native POSIX sockets or Winsock sockets, so Python is providing a very thin layer on top of those calls.
On both POSIX and Windows systems, this sequence of calls results in an OSError being raised. However, the location of the exception and the details of it are OS-specific. On POSIX systems, the exception is raised on the call to shutdown() and the errno value of the exception is set to 107 (Transport endpoint is not connected). On Windows systems, the exception is raised on the call to recvfrom() and the winerror value of the exception is set to 10038 (An operation was attempted on something that is not a socket). This means that there's no way to do this in an OS-agnositc way, the code has to account for both Windows and POSIX behavior and errors. Here's a simple example I wrote up:
import socket
import threading
import time
class MyServer(object):
def __init__(self, port:int=0):
if port == 0:
raise AttributeError('Invalid port supplied.')
self.port = port
self.socket = socket.socket(family=socket.AF_INET,
type=socket.SOCK_DGRAM)
self.socket.bind(('0.0.0.0', port))
self.exit_now = False
print('Starting server.')
self.thread = threading.Thread(target=self.run_server,
args=[self.socket])
self.thread.start()
def run_server(self, socket:socket.socket=None):
if socket is None:
raise AttributeError('No socket provided.')
buffer_size = 4096
while self.exit_now == False:
data = b''
try:
data, address = socket.recvfrom(buffer_size)
except OSError as e:
if e.winerror == 10038:
# Error is, "An operation was attempted on something that
# is not a socket". We don't care.
pass
else:
raise e
if len(data) > 0:
print(f'Received {len(data)} bytes from {address}.')
def stop(self):
self.exit_now = True
try:
self.socket.shutdown(socket.SHUT_RDWR)
except OSError as e:
if e.errno == 107:
# Error is, "Transport endpoint is not connected".
# We don't care.
pass
else:
raise e
self.socket.close()
self.thread.join()
print('Server stopped.')
if __name__ == '__main__':
server = MyServer(5555)
time.sleep(2)
server.stop()
exit(0)

Implement a quit command on the server and client sockets. Should work something like this:
Thread1:
status: listening
handler: quit
Thread2: client
exec: socket.send "quit" ---> Thread1.socket # host:port
Thread1:
status: socket closed()

To properly close a tcp socket in python, you have to call socket.shutdown(arg) before calling socket.close(). See the python socket documentation, the part about shutdown.
If the socket is UDP, you can't call socket.shutdown(...), it would raise an exception. And calling socket.close() alone would, like for tcp, keep the blocked operations blocking. close() alone won't interrupt them.
Many suggested solutions (not all), don't work or are seen as cumbersome as they involve 3rd party libraries. I haven't tested poll() or select(). What does definately work, is the following:
firstly, create an official Thread object for whatever thread is running socket.recv(), and save the handle to it. Secondly, import signal. Signal is an official library, which enables sending/recieving linux/posix signals to processes (read its documentation). Thirdly, to interrupt, assuming that handle to your thread is called udpThreadHandle:
signal.pthread_kill(udpthreadHandle.ident, signal.SIGINT)
and ofcourse, in the actual thread/loop doing the recieving:
try:
while True:
myUdpSocket.recv(...)
except KeyboardInterrupt:
pass
Notice, the exception handler for KeyboardInterrupt (generated by SIGINT), is OUTSIDE the recieve loop. This silently terminates the recieve loop and its thread.