Slow Rasa X in production

My rasa X server is running in production on the ubuntu machine for past one week. Initially, it works fine but if I don’t use it for like 2 3 days it starts responding to me very slow. Like it takes 2 3 minutes to log in only then update my dashboard for conversations etc. How to overcome this issue? do we need to refresh server everyday?

Capture

Hey @Abdur-Rub, how many conversataions do you have?

Just 4.

That’s definitely not right :thinking: Are there any indications in the logs as to why your rasa x and event services deployments are restarting? Maybe there are some clues there?

Let me share logs with you. I guess there are some issues. this is second time I facing this issue. I have public IP without DNS. is this okay?

These logs look fine – the first logs indicate that it seems that for some reason you are trying to POST directly the http://<ip>/api/ which isn’t a valid endpoint. Otherwise I can’t see any issues.

Has this happened multiple times? I’m wondering if it has something to do with your server rather than with rasa X.

I have simple socket.io webchat interface attached in html script. I faced this slow issue second time , first time it just simple start give me 404 page not found error, I then delete the cluster and make a fresh installation.

Hi @Abdur-Rub, can you please tell me where can I find the conversations users have with my bot through webchat ? I can’t find them on Rasa X conversations’ history.

Thanks a lot.