python - multiprocessing.Pool 在 Linux/Python2.7 上的 terminate() 之后生成新的子进程？

Question

我有一个可执行文件，我需要使用不同的参数经常运行它。为此，我使用多处理模块编写了一个小型 Python (2.7) 包装器，遵循给定的模式 here .

我的代码是这样的:

try:
     logging.info("starting pool runs")
     pool.map(run_nlin, params)
     pool.close()
 except KeyboardInterrupt:
     logging.info("^C pressed")
     pool.terminate()
 except Exception, e:
     logging.info("exception caught: ", e)
     pool.terminate()
 finally:
     time.sleep(5)
     pool.join()
     logging.info("done")

我的工作函数在这里:

class KeyboardInterruptError(Exception): pass

def run_nlin((path_config, path_log, path_nlin, update_method)):
    try:
        with open(path_log, "w") as log_:
            cmdline = [path_nlin, path_config]
            if update_method:
                cmdline += [update_method, ]
            sp.call(cmdline, stdout=log_, stderr=log_)
    except KeyboardInterrupt:
        time.sleep(5)
        raise KeyboardInterruptError()
    except:
        raise

path_config 是二进制程序配置文件的路径；里面有例如运行程序的日期。

当我启动包装器时，一切看起来都很好。但是，当我按下 ^C 时，包装器脚本似乎在终止之前从池中启动了一个额外的 numproc 进程。例如，当我在第 1-10 天启动脚本时，我可以在 ps aux 输出中看到二进制程序的两个实例正在运行(通常是第 1 天和第 3 天)。现在，当我按下 ^C 时，包装脚本退出，第 1 天和第 3 天的二进制程序消失了，但有新的二进制程序在第 5 天和第 7 天运行。

所以对我来说，似乎 Pool 在最终死亡之前启动了另一个 numproc 进程。

知道这里发生了什么，我能做些什么吗？

Answer 1

关于 this page ，多处理模块的作者 Jesse Noller 表明处理 KeyboardInterrupt 的正确方法是让子进程返回——而不是重新引发异常。这允许主进程终止池。

但是，如下面的代码所示，主进程不会到达 except KeyboardInterrupt block ，直到在 pool.map 已经运行。这就是为什么(我相信)在按下 Ctrl-C 之后，您会看到对辅助函数 run_nlin 的额外调用。

一种可能的解决方法是让所有工作函数测试是否已设置 multiprocessing.Event。如果事件已设置，则让 worker 提前退出，否则继续进行长计算。

---

import logging
import multiprocessing as mp
import time

logger = mp.log_to_stderr(logging.WARNING)

def worker(x):
    try:
        if not terminating.is_set():
            logger.warn("Running worker({x!r})".format(x = x))
            time.sleep(3)
        else:
            logger.warn("got the message... we're terminating!")
    except KeyboardInterrupt:
        logger.warn("terminating is set")        
        terminating.set()
    return x

def initializer(terminating_):
    # This places terminating in the global namespace of the worker subprocesses.
    # This allows the worker function to access `terminating` even though it is
    # not passed as an argument to the function.
    global terminating
    terminating = terminating_

def main():
    terminating = mp.Event()    
    result = []
    pool = mp.Pool(initializer=initializer, initargs=(terminating, ))
    params = range(12)
    try:
         logger.warn("starting pool runs")
         result = pool.map(worker, params)
         pool.close()
    except KeyboardInterrupt:
        logger.warn("^C pressed")
        pool.terminate()
    finally:
        pool.join()
        logger.warn('done: {r}'.format(r = result))

if __name__ == '__main__':
    main()

---

运行脚本会产生:

% test.py
[WARNING/MainProcess] starting pool runs
[WARNING/PoolWorker-1] Running worker(0)
[WARNING/PoolWorker-2] Running worker(1)
[WARNING/PoolWorker-3] Running worker(2)
[WARNING/PoolWorker-4] Running worker(3)

这里按下了 Ctrl-C；每个工作人员都设置了 terminating 事件。我们真的只需要一个来设置它，但尽管效率低下，但它仍然有效。

  C-c C-c[WARNING/PoolWorker-4] terminating is set
[WARNING/PoolWorker-2] terminating is set
[WARNING/PoolWorker-3] terminating is set
[WARNING/PoolWorker-1] terminating is set

现在 pool.map 排队的所有其他任务都已运行:

[WARNING/PoolWorker-4] got the message... we're terminating!
[WARNING/PoolWorker-2] got the message... we're terminating!
[WARNING/PoolWorker-1] got the message... we're terminating!
[WARNING/PoolWorker-2] got the message... we're terminating!
[WARNING/PoolWorker-4] got the message... we're terminating!
[WARNING/PoolWorker-2] got the message... we're terminating!
[WARNING/PoolWorker-1] got the message... we're terminating!
[WARNING/PoolWorker-3] got the message... we're terminating!

最后主进程到达 except KeyboardInterrupt block 。

[WARNING/MainProcess] ^C pressed
[WARNING/MainProcess] done: []