Problem using parallel computing in Python

Hello!

I have to prepare a Docker container with MCR and several Python packages (e.g. P1 and P2) compiled with SDK.

Both need parallel computing in functions (e.g. F1 and F2).

In Python, I import, initialize, run a function from the first one, then terminate.

Then I import the second one and when running a function from it, parallel pool is not started.

OS: Ubuntu 18.04, MATLAB 2019b

'local' profile used in both packages, selected in Preferences, local.mlsettings added to files of the packages.

Python (3.6.9) code:

import P1
h1=P1.initialize()
res1=h1.F1('params') 
# runs with parpool displaying "Starting parallel pool..", "Connected ..", "..shutting down" 
h1.terminate()
import P2
h2=P2.initialize()
res2=h2.F2('params')
# runs without parpool- messages not displayed
h2.terminate()

Is it possible to use parallel computing in this way or I have to compile all functions to one package (that is undesirable)?

Thanks, Ints