Rasa custom action on google colab/jupyter notebook doesn't work with chat(model_path, endpoints)?

I want to run both Rasa shell and Rasa custom action server on Google colab. Here are my settings, and part of code.

In the endpoints.yml.

action_endpoint:
  url: "http://localhost:5055/webhook"

actions.py

from typing import Any, Text, Dict, List

from rasa_sdk import Action, Tracker
from rasa_sdk.executor import CollectingDispatcher

class ActionHelloWorld(Action):

     def name(self) -> Text:
         return "action_hello_world"

     def run(self, dispatcher: CollectingDispatcher,
             tracker: Tracker,
             domain: Dict[Text, Any]) -> List[Dict[Text, Any]]:

         dispatcher.utter_message(text="Hello World from action!")

         return []

After running Action server in the background using nohup.

!nohup /usr/bin/python3 -m rasa_sdk.endpoint --actions actions &

I trained the Rasa model using :

model_path = rasa.train(domain, config, [training_files] , output)

Then, I run the Rasa shell in colab cell using :

from rasa.jupyter import chat

endpoints = "endpoints.yml"

chat(model_path, endpoints)

This method somehow doesn’t connect with the Rasa action server, it will work for non-action replies. it just doesn’t work for getting custom replies.

But if I run the Rasa shell using !rasa shell --endpoints endpoints.yml. Both part can be seen interacting but they won’t if chat(model_path, endpoints) is used ?