n8n responds with “Please provide your city, or something else” and must remember the context THAT FROM NOW WE WAIT FOR THIS PERSONAL INFO - this is the main issue
user send it’s city - for Example London
n8n retrieves context, understands that it is waiting the city name. Then it get city name and query me backend to register user
so the main issue is how to keep context and distinguish the regular messages (I already have this flow to respond to user via open ai agent) from cases when it exactly waits for city from the user input
We’ve created a new category for help with designing workflows, and I’ve moved your question there: Help me Build my Workflow. Find out how this category works by reading this topic.
Hi! Thank you! What is the best choice of database you would recommend? I guess that google sheet has a high io time.
Maybe there are some in memory caches (like redis or memcached) from scratch. It’s worth to mention that this database call will arise on every telegram message
Depends on your needs/setup. If you’re on n8n cloud and you are going to receive massive amounts of messages 3k+ a day, it might be a good idea to look at self hosting as the execution cost can increase a lot.
If you’re only receiving a modest mount of messages and are on cloud I like supabase.
The other advantage of self host is you can host a Postgres DB locally. And yes if you wanted to add a caching layer like redis that is possible.
This would probably be most easily completed with a docker compose file.