What’s the preferred way to deploy to Rasa X with a custom connector?
I have a Facebook Messenger bot that handles incoming images from users. When I was testing locally I manipulated Rasa’s facebook.py file to handle incoming images and I have no issues.
Now I’m looking to deploy the bot to a server and I’m unsure how to go about deploying my bot with this custom connector.
I don’t have a ton of experience with deployment tasks so any help is greatly appreciated!
Thanks for the response! I worked at this for a couple days and think I understand the docker-compose deployment process a little better now. I was able to setup the custom connector on my EC2 instance with the following steps. Do these appear correct? Or is there a better way to go about accomplishing this?
In the specific case here I copied the default facebook.py connector then added some code to pass images to my bot, which is the custom_facebook.py file. So the same functionality of the default facebook connector is maintained, just extended a little bit to fit my use case.
I haven’t tried adding other channels, but if you setup additional channels in credentials.yml and connected the webhooks I don’t see why that wouldn’t work.
have been trying to get Alexa custom connectors working on prod, but was no luck since 2 months. Your steps magically started working, thanks for this post saved my time.
@Tobias_Wochinger
On the Rasa X conversations tab, do all conversations by different Users with the custom channel (Facebook/Alexa) appear in different tabs or one single tab
asking for Issue #45839
yep. Different conversation IDs get different users. Note that depending on the channel connector a user might have different conversation IDs (they should be the same with Facebook / Alexa though)