从 main 终止多处理进程

Question

I am using Redis Pubsub to trigger a child process and save a reference to it. I would like to terminate the previous process when the next message comes in. Unfortunately, though I can see the child process in debugger, and it has a terminate() function, it does not seem like main can see it-I get an error saying 'NoneType' object has no attribute 'terminate'. Is there a straightforward way to terminate the process?

My code (in ' main ':

conn = redis.Redis(host="localhost", port="6379")
if not conn.ping():
        raise Exception('Redis unavailable')

pubsub = conn.pubsub()
pubsub.subscribe("feed")
data = None
loaderProcess = None

for message in pubsub.listen():
    logging.info("received pubsub message")
    logging.info(message)
    logging.info(message['type'])
    if message['type'] == "message":
        data = json.loads(message.get("data"))
        if data and data['source']:
            try:
                if loaderProcess is not None:
                    loaderProcess.terminate()
                    loaderProcess.join()
                args.infile = data['source']
                loader = Video(infile=data.get("source"), fps=30.0)
                loaderProcess = multiprocessing.Process(target=load, args = (loader, conn, args,))                    
            except Exception as e:
                logging.error("Error occurred", exc_info=True)

Stack trace:

ERROR:root:Error occurred
Traceback (most recent call last):
  File "C:\video-analysis\capture.py", line 140, in <module>
    loaderProcess.terminate()
  File "C:\Users\bkogan\AppData\Local\Programs\Python\Python39\lib\multiprocessing\process.py", line 133, in terminate
    self._popen.terminate()

Answer 1

OK, got it. A few issues. One was that I was not starting the process explicitly. Another is that my reference to the process was not working. Adding the process to a list works, though:

procs = []

for message in pubsub.listen():
    try:
        if message['type'] == "message":
            data = json.loads(message.get("data"))
            if data and data['source']:
                    for proc in procs:
                        if proc.is_alive():
                            proc.terminate()
                            proc.join(timeout=0)
                            procs.pop(0)
                    loaderProcess = multiprocessing.Process(target=load, args = (data.get("source"), args,))
                    procs.append(loaderProcess)
                    loaderProcess.start()
                    continue
    except Exception as e:
        logging.error("Error occurred", exc_info=True)