Hi Team,
We are stuck with a issue, we are trying to use multiprocessing. We are using below ways.
1. from multiprocessing.dummy import Pool
pool = Pool(20)
pool.apply_async(call_api, xxxxxxxx)
2. import concurrent.futures
import requests
pool = concurrent.futures.ThreadPoolExecutor(max_workers=4)
We are using python 3.7 and app.yaml file is below.
service: abc
runtime: python37
instance_class: B4_1G
entrypoint: gunicorn -w 4 -k uvicorn.workers.UvicornWorker app.main:app
basic_scaling:
idle_timeout: 10m
max_instances: 25
Both the methods are working fine in our local machines(computers) but when deploying on app engine these functionality are not working even we are not getting any error in logs.
Our use case is, we have two services A and B in service B we have written a api to upload file into google bucket, and we are calling this api from service A, we are uploading big files so we cant wait to complete, so we want response back from service A after call Service B api.
Please help to understand, what is the issue and how can we achieve this.