I try to use pathos.multiprocessing.Pool in my project.
However, it will meet the following problem when I terminate the Pool.
I use CentOS 6.5, I'm not sure if it is caused by pathos.multiprocessing.Pool or other thing, can anyone help me on it?
Traceback (most recent call last):
File "/usr/local/lib/python2.7/threading.py", line 801, in __bootstrap_inner
self.run()
File "/usr/local/lib/python2.7/threading.py", line 1073, in run
self.function(*self.args, **self.kwargs)
File "receiver.py", line 132, in kill_clients
pool.terminate()
File "/usr/local/lib/python2.7/site-packages/multiprocess/pool.py", line 465, in terminate
self._terminate()
File "/usr/local/lib/python2.7/site-packages/multiprocess/util.py", line 207, in __call__
res = self._callback(*self._args, **self._kwargs)
File "/usr/local/lib/python2.7/site-packages/multiprocess/pool.py", line 513, in _terminate_pool
p.terminate()
File "/usr/local/lib/python2.7/site-packages/multiprocess/process.py", line 137, in terminate
self._popen.terminate()
File "/usr/local/lib/python2.7/site-packages/multiprocess/forking.py", line 174, in terminate
os.kill(self.pid, signal.SIGTERM)
OSError: [Errno 3] No such process
The wired thing is that at the beginning, it works well. But when the 4th job is received, there will be such problem.
class Receiver:
def __init__(self):
....
self.results={}
def kill_clients(self, client_list, pool):
for client in client_list:
client.kill()
pool.terminate()
def process_result(self, result):
if result is None:
self.results = {}
return
res = result.split(':')
if len(res) != 4:
raise Exception("result with wrong format: %s" % result)
self.results['%s_%s' % (res[0], res[1])] = {"code": res[3], "msg": res[4]}
...
def handler(self, job):
self.lg.debug("Receive job in rtmp_start_handler.")
self.lg.debug("<%s>" % str(job))
# each client corresponding one process
cli_counts = job['count']
pool = Pool(processes=cli_counts)
clients = []
try:
for i in xrange(cli_counts):
rtmp_cli = RtmpClient(job['case'], i)
clients.append(rtmp_cli)
[pool.apply_async(client.run, callback=self.process_result)
for client in clients]
pool.close()
sleep(1)
self.lg.debug("All clients are started.")
t = Timer(
job['timeout'],
self.kill_clients,
args=(clients, pool)
)
t.start()
self.lg.debug("Timer is started. timeout %s s" % job['timeout'])
pool.join()
except Exception, e:
self.lg.warning("Exception occurred: %s" % e)
self.lg.warning(format_exc())
return "0"
# here the self.results shall be ready
return self.parse_results()
The OSError is not caused by the Pool but by my program issue.
When I use Popen to create a subprocess and exec ffmpeg, it will exit immediately(due to other problem), so when I try to kill the subprocess, it it not existed by then. That's why OSError will be raised.
Related
I am trying to multithread a Singal R connection with python, I am getting a connection Rawlink error the connections works and I need to wait 1 second in order to receive the message from the client, I am also using a barrier to execute the thread "simultaneous".
Here is my code
with Session() as session:
global connection
connection = Connection("http://sampleSINGALRURL/signalr", session)
presenceservice = connection.register_hub('ClientRegistration')
presenceservice1 = connection.register_hub('PresenceClientHub')
connection.start()
presenceservice.server.invoke('IdentifyClient', devideIdentity, softwareVersion, IpAddress,
machineName, DeviceType, patientAdmissionGuid, patientID, pairingId)
presenceservice1.client.on('StaffPresenceNotified', self.get_data1)
connection.wait(1)
And then my threading functions
def get_clients(self):
global barrier
self.connect_to_database1()
barrier.wait()
self.get_message_from_client1()
self.print_data1()
def send_messages(self):
global MessageNumber
global machineName
global staffName
global request
machineName = final_result[MessageNumber][0]
staffName = staff_results[MessageNumber][0]
MessageNumber += 1
barrier.wait()
request = requests.post(
"http://sampleurl/api/sample")
return request
def print_response(self):
global request
timestamp = datetime.utcnow().strftime('%Y-%m-%d %H:%M:%S.%f')[:-3]
logging.info("Message sent at " + "Time : " + timestamp + " " + machineName)
def Spin_Clients(self, NumMessages):
for i in range(10):
self.client_list.append(Thread(target=self.send_messages))
self.client_list[i].start()
self.print_response()
sleep(2)
for i in range(10):
self.Message_List.append(Thread(target=self.get_clients))
self.Message_List[i].start()
for thread in self.client_list:
thread.join()
for thread in self.Message_List:
thread.join()
Error logs
All threads have finished
11:41:37.243
Exception in thread Thread-13: Traceback (most recent call last): File "c:\users\appdata\local\programs\python\python37\lib\threading.py", line 917, in _bootstrap_inner
self.run()
File "c:\users\appdata\local\programs\python\python37\lib\threading.py", line 865, in run
self._target(*self._args, **self._kwargs)
File "C:\Data\LoadTest.py", line 103, in get_clients
self.get_message_from_client1()
File "C:\Data\LoadTest.py", line 89, in get_message_from_client1
connection.wait(1)
File "c:\users\appdata\local\programs\python\python37\lib\site-packages\signalr\_connection.py", line 60, in wait
gevent.joinall([self.__greenlet], timeout)
File "src\gevent\greenlet.py", line 849, in gevent._greenlet.joinall
File "src\gevent\greenlet.py", line 859, in gevent._greenlet.joinall
File "src\gevent\_hub_primitives.py", line 198, in gevent.__hub_primitives.wait_on_objects File "src\gevent\_hub_primitives.py", line 235, in gevent.__hub_primitives.wait_on_objects
File "src\gevent\_hub_primitives.py", line 125, in gevent.__hub_primitives._WaitIterator.__iter__ AttributeError: 'NoneType' object has no attribute 'rawlink'
I also tried to use locks but that had the same outcome.
Any idea?
I am trying to use my other cores in my python program. And the following is the basic structure/logic of my code:
import multiprocessing as mp
import pandas as pd
import gc
def multiprocess_RUN(param):
result = Analysis_Obj.run(param)
return result
class Analysis_Obj():
def __init__(self, filename):
self.DF = pd.read_csv(filename)
return
def run_Analysis(self, param):
# Multi-core option
pool = mp.Pool(processes=1)
run_result = pool.map(multiprocess_RUN, [self, param])
# Normal option
run_result = self.run(param)
return run_result
def run(self, param):
# Let's say I have written a function to count the frequency of 'param' in the target file
result = count(self.DF, param)
return result
if __name__ == "__main__":
files = ['file1.csv', 'file2.csv']
params = [1,2,3,4]
results = []
for i in range(0,len(files)):
analysis = Analysis_Obj(files[i])
for j in range(0,len(params)):
result = analysis.run_Analysis(params[j])
results.append(result)
del result
del analysis
gc.collect()
If I comment out the 'Multi-core option' and run the 'Normal option' everything runs fine. But even if I run the 'Multi-core option' with processes=1 I get a Memory Error when my for loop starts on the 2nd file. I have deliberately set it up so that I create and delete an Analysis object in each for loop, so that the file that has been processed will be cleared from memory. Clearly this hasn't worked. Advice of how to get around this would be very much appreciated.
Cheers
EDIT:
Here is the error message I have in the terminal:
Exception in thread Thread-7:
Traceback (most recent call last):
File "/usr/lib/python2.7/threading.py", line 801, in __bootstrap_inner
self.run()
File "/usr/lib/python2.7/threading.py", line 754, in run
self.__target(*self.__args, **self.__kwargs)
File "/usr/lib/python2.7/multiprocessing/pool.py", line 326, in _handle_workers
pool._maintain_pool()
File "/usr/lib/python2.7/multiprocessing/pool.py", line 230, in _maintain_pool
self._repopulate_pool()
File "/usr/lib/python2.7/multiprocessing/pool.py", line 223, in _repopulate_pool
w.start()
File "/usr/lib/python2.7/multiprocessing/process.py", line 130, in start
self._popen = Popen(self)
File "/usr/lib/python2.7/multiprocessing/forking.py", line 121, in __init__
self.pid = os.fork()
OSError: [Errno 12] Cannot allocate memory
Is it possible to spawn multiple processes from a single thread? Or is it a proper design to implement?
My code sample is -
def run_all_tasks(self):
for platform in self._platforms:
task_thread = threading.Thread(
target=self._run_task_list, args=(
self._get_tasks(),platform,))
taskset_threads.append(task_thread)
for taskset_thread in taskset_threads:
taskset_thread.start()
for taskset_thread in taskset_threads:
taskset_thread.join()
def _run_task_list(self, tasklist, platform):
try:
test_case_name = task.__class__.__name__
try:
test_case_name = task._get_test_case_name()
except:
test_case_name = task.__class__.__name__
pass
max_runtime = task.get_max_runtime()
manager = Manager()
self._shared_mem = manager.dict()
for task in tasklist:
task_proc = Process(
target=self.proc_setup,
args=(task, self, self._shared_mem))
task_proc.start()
task_proc.join(max_runtime)
This works however, sometimes it gives following error -
Traceback (most recent call last):
File "C:\wor\lib\TaskSet.py", line 430, in _run_task_list
if "warning" in self._shared_mem:
File "<string>", line 2, in __contains__
File "C:\Python27\lib\multiprocessing\managers.py", line 755, in _callmethod
self._connect()
File "C:\Python27\lib\multiprocessing\managers.py", line 742, in _connect
conn = self._Client(self._token.address, authkey=self._authkey)
File "C:\Python27\lib\multiprocessing\connection.py", line 167, in Client
c = PipeClient(address)
File "C:\Python27\lib\multiprocessing\connection.py", line 387, in PipeClient
win32.WaitNamedPipe(address, 1000)
WindowsError: [Error 2] The system cannot find the file specified
This can also be seen on linux platform.
I have a python program like below.
from multiprocessing import Lock, Process, Queue, current_process
import time
lock = Lock()
def do_job(tasks_to_accomplish, tasks_that_are_done):
while not tasks_to_accomplish.empty():
task = tasks_to_accomplish.get()
print(task)
lock.acquire()
tasks_that_are_done.put(task + ' is done by ' + current_process().name)
lock.release()
time.sleep(1)
return True
def main():
number_of_task = 10
number_of_processes = 4
tasks_to_accomplish = Queue()
tasks_that_are_done = Queue()
processes = []
for i in range(number_of_task):
tasks_to_accomplish.put("Task no " + str(i))
# creating processes
for w in range(number_of_processes):
p = Process(target=do_job, args=(tasks_to_accomplish, tasks_that_are_done))
processes.append(p)
p.start()
# completing process
for p in processes:
p.join()
# print the output
while not tasks_that_are_done.empty():
print(tasks_that_are_done.get())
return True
if __name__ == '__main__':
main()
Sometimes program run perfectly but sometimes it gets stuck and doesn't complete. When quit manually, it produces following error.
$ python3 multiprocessing_example.py
Task no 0
Task no 1
Task no 2
Task no 3
Task no 4
Task no 5
Task no 6
Task no 7
Task no 8
Task no 9
^CProcess Process-1:
Traceback (most recent call last):
File "multiprocessing_example.py", line 47, in <module>
main()
File "multiprocessing_example.py", line 37, in main
p.join()
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/multiprocessing/process.py", line 121, in join
res = self._popen.wait(timeout)
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/multiprocessing/popen_fork.py", line 51, in wait
return self.poll(os.WNOHANG if timeout == 0.0 else 0)
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/multiprocessing/popen_fork.py", line 29, in poll
pid, sts = os.waitpid(self.pid, flag)
KeyboardInterrupt
Traceback (most recent call last):
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/multiprocessing/process.py", line 249, in _bootstrap
self.run()
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/multiprocessing/process.py", line 93, in run
self._target(*self._args, **self._kwargs)
File "multiprocessing_example.py", line 9, in do_job
task = tasks_to_accomplish.get()
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/multiprocessing/queues.py", line 94, in get
res = self._recv_bytes()
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/multiprocessing/connection.py", line 216, in recv_bytes
buf = self._recv_bytes(maxlength)
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/multiprocessing/connection.py", line 407, in _recv_bytes
buf = self._recv(4)
File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/multiprocessing/connection.py", line 379, in _recv
chunk = read(handle, remaining)
KeyboardInterrupt
Can someone tell me what is the issue with the program? I am using python 3.6.
Note: Lock is not needed around a Queue.
lock.acquire()
tasks_that_are_done.put(task + ' is done by ' + current_process().name)
lock.release()
Queue
The Queue class in this module implements all the required locking semantics.
Question: ... what is the issue with the program?
You are using Queue.empty() and Queue.get(),
such leads to Deadlock on calling join() because there is no guarantee that the empty() State don't change until get()
was reaching.
Deadlock prone:
while not tasks_to_accomplish.empty():
task = tasks_to_accomplish.get()
Instead of using empty/get, Pair use for instance:
import queue
while True:
try:
task = tasks_to_accomplish.get_nowait()
except queue.Empty:
break
else:
# Handle task here
...
tasks_to_accomplish.task_done()
I have a Python script that uses fabric library to test SSH connection to multiple hosts. I want to gather all the results in one list:
...
import fabric
from fabric.api import *
results = []
#parallel
def test_connection():
global results
try:
run('ls')
results += "%s: SUCCESS" % env.host
except Exception as e:
results += "%s: FAILURE. Exception: %e" % (env.host, e)
if __name__ == '__main__':
tasks.execute(test_connection)
print results
When I execute the script, I get the following:
Traceback (most recent call last):
File "./test_ssh.py", line 99, in <module>
tasks.execute(test_connection)
File "/Library/Python/2.7/site-packages/fabric/tasks.py", line 387, in execute
multiprocessing
File "/Library/Python/2.7/site-packages/fabric/tasks.py", line 277, in _execute
return task.run(*args, **kwargs)
File "/Library/Python/2.7/site-packages/fabric/tasks.py", line 174, in run
return self.wrapped(*args, **kwargs)
File "./test_ssh.py", line 96, in test_connection
results += "%s: FAILURE. Exception: %e" % (env.host, e)
UnboundLocalError: local variable 'results' referenced before assignment
I think it is because test_connection runs it's own context so it doesn't have access to results.
Is there another way I can gather my results then ?
The trick is that you can actually return results from parallel execution:
#parallel
def test_connection():
try:
run('ls')
return True
except Exception:
return False
Now when you call your task, you will get:
result = execute(test_connection)
results = [ ('HOST %s succeeded' % key) if value else ('HOST %s failed' % key) for key, value in result.items() ]