python - SQLAlchemyJobstore 在 Flask 应用程序中返回 'TypeError: cannot pickle 'SSLContext' object'

Question

在我的代码中实现 SQLAlchemyJobStore 后出现问题。尝试添加作业会导致以下错误：TypeError: cannot pickle 'SSLContext' object。我已经阅读了所有文档并尝试了类似问题提到的所有修复但没有结果（例如https://github.com/agronholm/apscheduler/issues/421）。

我的代码如下：

在views.py中：

@app.before_first_request
def schedule_jobs():
 
    scheduler = BackgroundScheduler(
        jobstores={'default': SQLAlchemyJobStore(url='sqlite:///jobs.sqlite')}
    )
    scheduler.start()
    data = Data()
    scheduler.add_job(data.schedule_app,
                      trigger='interval',
                      hours=int(os.getenv('UPDATE_FREQUENCY')))

在 data.py 中：

class Data:

    def __init__(self):
        self.ssl_context = ssl.SSLContext()

    def __getstate__(self):
        state = self.__dict__.copy()
        del state['ssl_context']
        return state

    def __setstate__(self, state):
        self.__dict__.update(state)
        self.ssl_context = ssl.SSLContext()

    def schedule_app(self):
        print("Task added Successfully")

我添加了setstate和getstate作为上述修复类似问题的一部分。即使在课堂上只有 schedule_app，我仍然得到相同的结果。

完整的跟踪：

2021-04-03T10:18:48.206620307Z Traceback (most recent call last):
2021-04-03T10:18:48.206623307Z   File "/tmp/8d8f6862c98e628/antenv/lib/python3.8/site-packages/flask/app.py", line 2447, in wsgi_app
2021-04-03T10:18:48.206626507Z     response = self.full_dispatch_request()
2021-04-03T10:18:48.206629507Z   File "/tmp/8d8f6862c98e628/antenv/lib/python3.8/site-packages/flask/app.py", line 1945, in full_dispatch_request
2021-04-03T10:18:48.206632707Z     self.try_trigger_before_first_request_functions()
2021-04-03T10:18:48.206635807Z   File "/tmp/8d8f6862c98e628/antenv/lib/python3.8/site-packages/flask/app.py", line 1993, in try_trigger_before_first_request_functions
2021-04-03T10:18:48.206638907Z     func()
2021-04-03T10:18:48.206641807Z   File "/tmp/8d8f6862c98e628/app/views.py", line 39, in schedule_jobs
2021-04-03T10:18:48.206644907Z     scheduler.add_job(data.schedule_app,
2021-04-03T10:18:48.206647807Z   File "/tmp/8d8f6862c98e628/antenv/lib/python3.8/site-packages/apscheduler/schedulers/base.py", line 448, in add_job
2021-04-03T10:18:48.206650907Z     self._real_add_job(job, jobstore, replace_existing)
2021-04-03T10:18:48.206653807Z   File "/tmp/8d8f6862c98e628/antenv/lib/python3.8/site-packages/apscheduler/schedulers/base.py", line 872, in _real_add_job
2021-04-03T10:18:48.206656907Z     store.add_job(job)
2021-04-03T10:18:48.206659807Z   File "/tmp/8d8f6862c98e628/antenv/lib/python3.8/site-packages/apscheduler/jobstores/sqlalchemy.py", line 95, in add_job
2021-04-03T10:18:48.206662907Z     'job_state': pickle.dumps(job.__getstate__(), self.pickle_protocol)
2021-04-03T10:18:48.206665907Z TypeError: cannot pickle 'SSLContext' object

有没有办法来解决这个问题？

我使用 SQLAlchemyJobStore 只是为了避免多个实例中的多个工作人员执行计划的作业（我在 Azure 中使用 8 个实例运行此应用程序并遵循 Gunicorn 配置：

gunicorn -w 8 --preload --worker-class gevent

有什么方法可以在不使用 SQLAlchemyJobStore 的情况下实现这一目标？

先感谢您。

标签： pythonazureflasksqlalchemyapscheduler