使用并发期货 ProcessPoolExecutor 时登录主进程

Question

关于如何让分叉进程登录到主进程有什么想法吗？

我的设置是这样的：

from itertools import repeat
from concurrent.futures import ProcessPoolExecutor

logger = logging.getLogger(__name__)

def worker_process(i, logger):
    logger.info(f"{i}")

with ProcessPoolExecutor(max_workers=4) as executor:
    logger.info("Starting")
    executor.map(worker_process, range(5), repeat(logger))

Answer 1

您的工作进程确实记录到主线程。您没有看到任何日志输出，因为 logger 的默认日志级别高于 INFO。

将日志记录设置为 INFO，您将看到输出：

from itertools import repeat
from concurrent.futures import ProcessPoolExecutor
import logging

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)


def worker_process(i, logger):
    logger.info(f"{i}")


with ProcessPoolExecutor(max_workers=4) as executor:
    executor.map(worker_process, range(5), repeat(logger))

结果：

INFO:__main__:0
INFO:__main__:4
INFO:__main__:3
INFO:__main__:1
INFO:__main__:2

使用并发期货 ProcessPoolExecutor 时登录主进程

Logging in main process when using concurrent futures ProcessPoolExecutor

python

python-3.x

logging

multiprocessing

concurrent.futures