**exe.py**
def createProcess(f):
try:
from multiprocessing import Process
newProcess = Process(target=f)
newProcess.start()
newProcess.join()
except:
print "Error creating process"
def lala():
print "success creating process"
print "tying to make a process"
from multiprocessing import Process
newProcess = Process(target=lala)
newProcess.start()
**main.py**
if __name__ == '__main__':
f = open("exe.py", "r")
b = f.read()
f.close()
o = compile(b, "exe.py", "exec")
eval(o)
我收到以下错误
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "C:\Program Files\Opsware\agent\lcpython15\lib\multiprocessing\forking.p
", line 374, in main
self = load(from_parent)
File "C:\Program Files\Opsware\agent\lcpython15\lib\pickle.py", line 1378, in
load
return Unpickler(file).load()
File "C:\Program Files\Opsware\agent\lcpython15\lib\pickle.py", line 858, in
oad
dispatch[key](self)
File "C:\Program Files\Opsware\agent\lcpython15\lib\pickle.py", line 1090, in
load_global
klass = self.find_class(module, name)
File "C:\Program Files\Opsware\agent\lcpython15\lib\pickle.py", line 1126, in
find_class
klass = getattr(mod, name)
AttributeError: 'module' object has no attribute 'lala'
稍后编辑
我将exe.py更改为
def lala2():
f = open("C:\\work\\asdfas", "w")
f.write("dsdfg\r\n")
f.close()
print "success creating process"
if __name__ == '__main__':
print "tying to make a process"
from multiprocessing import Process, freeze_support
freeze_support()
import pickle
l = pickle.dumps(lala2)
pickle.loads(l)()
newProcess = Process(target=pickle.loads(l))
newProcess.daemon = True
newProcess.start()
if newProcess.is_alive() == True:
print "alive"
else:
print "not alive"
import time
time.sleep(12)
这应该可以导入它,并且由于pickle测试它显示我的方法是可选择的。关于它为何如此行事的任何建议?
答案 0 :(得分:2)
你在Windows上。不幸的是,在Windows上,不可能将动态代码对象用作multiprocessing
的目标,因为多处理的Windows实现必须能够导入主模块(技术原因是因为Windows缺少本机{{ 1}}等价物)。有关适用限制的详细信息,请参阅multiprocessing programming guidelines for Windows。
解决方案是将代码写入文件,然后fork()
将其写入服务器。