Flask - 无尽的后台进程

Question

I'm working on flask project to collect measurement data from IOT sensors in database. Application have endopoints for sensors and for frontend. Currently I have to add endless background process, that check some data in database like communication timeouts, exceeding the alarm levels etc. and send sms/emails alerts. What is the proper way to do that. Thread is tempting for me, but I wonder how to avoid multiple threads when I run multiple workers with gunicorn. Of course i want to use all my DB models and flask-sqlalchemy. I read Miguel Grinberg blog about Background Jobs but I have doubts about it is the proper way to resolve my problem.

Answer 1

An event based approach is far better than a polling approach (endless répétition of the same task to check for some condition). It would preferable to check data before storing them in the data base or executing a task to check data upon an update of the data base. As for timeout an error should occur and should be related to an exception that would run a fallback task. To summarize use an event based pattern to check rather a continuous backgroud task.