最近一个项目上使用线程池，设定处理项1W，10线程，但是需要检测线程进行状态。出现错误N次，就自动终止线程。实现如下:

线程池代码

ERRORCOUNT=0

IS_EXIT=True

class Worker(Thread):

worker_count=0

timeout=1

def __init__(self, workQueue,resultQueue,**kwds):

Thread.__init__(self,**kwds)

self.id=Worker.worker_count

Worker.worker_count+=1

self.setDaemon(True)

self.workQueue=workQueue

self.resultQueue=resultQueue

self.start()

def run(self):

#the get-some-work,do-some-work main loop of worker threads

while IS_EXIT:

try:

callable,args,kwds=self.workQueue.get(timeout=Worker.timeout)

res=callable(*args,**kwds)

self.resultQueue.put(res)

except Queue.Empty:

break

except:

pass

class WorkerManager:

def __init__(self, num_of_workers=10,timeout=2):

self.workQueue=Queue.Queue()

self.resultQueue=Queue.Queue()

self.workers=[]

self.timeout=timeout

self._recruitThreads(num_of_workers)

def _recruitThreads(self,num_of_workers):

for i in range(num_of_workers):

worker=Worker(self.workQueue,self.resultQueue)

self.workers.append(worker)

def wait_for_complete(self):

#then,wait for each of them to terminate

while len(self.workers):

worker=self.workers.pop()

worker.join(10)

if worker.isAlive() and not self.workQueue.empty():

self.workers.append(worker)

def add_job(self,callable,*args,**kwds):

self.workQueue.put((callable,args,kwds))

def get_result(self,*args,**kwds):

return self.resultQueue.get(*args,**kwds)

在WorkerManager中使用到一个全局变量IS_EXIT用来判断是否需要退出线程

调用线程代码

wm=WorkerManager(10)#10线程

for i in range(10000):

wm.add_job(do_job)

wm.wait_for_complete()

具体工作代码

def do_job():

global ERRORCOUNT

global IS_EXIT

try:

do anything

except:

ERRORCOUNT+=1

if ERRORCOUNT>5:

IS_EXIT=False

此处使用了全局变量ERRORCOUNT统计错误数量，超过指定次数，则设置IS_EXIT=False通知线程停止执行。

至此基本上满足项目所需，但并不友好，应有更好的方式。