but after the auth i don't seem to receive any cookies to the server

Hi guys

I want to know how BaaS like Supabase and Firebase works? I want to try to make one

You create dynamic database instances and expose an API to do so, as well as map SQL commands to API calls For realtime things get more complicated as you need your server to also work as websocket broker between the API/DB and the clients, and you need to listen to database changes so you can broadcast those into the connected websockets, and vise versa. (in a nutshell)

Is there any small project available open source?or any tutorial?

google showed this: https://blog.back4app.com/baas-open-source/

What's the hardware being used? like how many physical cores and ram is being allocated to my database or can i get a approximate aws instance like t2.micro

Am in the free tier if that matters

Guys any feedback on refine.dev Looks great.. Has anyone used it?

nope, hearing it for the first time.

Do projects enter a "sleep mode" because the first API call I do after some hours of idleness always takes substantially more to arrive than subsequent calls.

Never faced this issue

I've noticed email confirmations for signups sometimes take several minutes, sometimes 10 or 15. Is there a way to speed this up?

You can speed this up by using Custom SMTP

Any docs on that part? Sendgrid seems to change every week.

Is that a self hosted feature not available otherwise?

Go to Authentication/Settings in your dashboard then scroll down to Enable Custom SMTP. It's as easy as that.

Thank you!

Hi, I just signed up for supabase, I see about RAM on database. Does it mean I am limited by RAM ? right now it's sitting on 40%, realistically how many operation can I do before reaching 100% RAM

I intend to migrate my firestore project to supabase, but in firestore I don't have to deal with RAM. although in supabase, I don't have to deal with api calls count. Anyone can explain how this RAM works ?

I have a table with Blog Posts. Aauthenticated users can Favorite and Upvote blog posts. What would be the best practice to store these Favorite and Upvotes? So that it makes it easy to use the SDK and that is scaleable. I come from Firebase and I would just update a Favorites array in a specific Blog Post row with the user id (not even sure if that would considered best practice). I'm kinda stuck because I want to set this up right. Thanks !

Anyone use supabase in production for about 10k read / write in short period of time (e.g. 1 minute) ? can it scale ? I come from firestore, so I don't have to worry about this. Should I worried using supabase ? looks like they have limit on memory / cpu usage.

167 requests per second shouldn't be a challenge for any DB or DBaaS out there. There's potentially hundreds of factors which affect Postgres performance - Do you use indexes? Do you have partitions? Do you have suitable primary keys? Are you querying views from inside views? Are you hosting in the same region as your front-end?

@User no, I don't use any complicated query / stuff. it's a very simple query, with exact string id.

Can it scale to let's say 20k, 30k or even 100k?

Love is when you type "s" in the browser address bar and it automatically suggests supabase.io as the first option

Supabase scaling

You can create a number field and simply decrease and increase. That is the easiest solution. The second easiest is to create an upvote table and add a primary key relation to the post. Relations are handled fairly easily in the SDK

Noted. Reason I’m struggling is because the Upvotes are for logged in users only. Would you say it’s best to create an Upvote table, or store Upvotes in the User or BlogPost table? So many options :S

I would do a upvotes table