Hi, I want to ask for a question: I’m deploying a Rasa chatbot on a server, and I want to put it on production. The problem is that, say, I want to update my Rasa-core model because I’ve added a new intent or slot. Then, I have to retrain, shut down my server and run it again, but when it’s off I’m without chatbot to answer to the clients. Is there any way to handle this? I was thinking in a online model updating or in two parallel servers, where first one of them it’s updated (and the queries are sent to the other one) and then the second one is updated, but in this case I would need to share the tracker between the two servers. Could anyone help me please? Thanks!!

Share tracker between two servers

basque21 (Basque21) October 23, 2018, 10:53am 7

solved! thanks!

Topic		Replies	Views
Duplicate key url when setting up tracker store (docker) Rasa Open Source	1	501	January 29, 2020
Add conversation persistence with load balancer (nginx, redis, found possible solution) Rasa Open Source	1	616	July 29, 2022
Save tracker in redis Rasa Open Source	5	964	January 2, 2019
Multi-conversation Rasa Core with tracker store Rasa Open Source	1	854	April 19, 2019
Trying to use redis tracker store with rasa_core.server Rasa Open Source	3	834	January 4, 2019