Issue when running RASA with multiple SANIC_WORKERS

We have the RASA and ACTION service Deployed in Kubernetes in as separate service. We are running RASA with 5 SANIC_WORKERS.

When there is single SANIC_WORKER, different sessions are handled and their is no issue in that. But with multiple SANIC_WORKER, when different request are made within the same session, the slot values are not persisting, i.e., when the initial request is made, it is handled by a single sanic worker and we are able to get the slot value but when second request is made within the same session, If that request is handled by other sanic worker, we are getting NULL or NONE slot value.

Below is the action code where we are getting null value :

class SRImoreInfo(Action):

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


def run(self, dispatcher: CollectingDispatcher,
        tracker: Tracker,
        domain: Dict[Text, Any]) -> List[Dict[Text, Any]]:
    user_id = tracker.get_slot("user_id")
    user_link = credentials.server_url + str(user_id)
    print("user_link", user_link)
    dispatcher.utter_message(text = "[Click Here](" + user_link + ") for more information.")
    return []

In the above code, we are receiving null value in tracker.get_slot(“user_id”) when running RASA with 5 SANIC_WORKER.

Please share your thoughts what wrong am I doing? Let me know If any further information is required for better understanding. Any sort of suggestion will be appreciated.

Which tracker- and lock store are you using? Multiple workers dont work with in memory stores