Hi guys, I am not sure if my assumption is correct Graphql server sends back whatever clients requests. In order to do so they would need to query the database for the exact keys to fetch. For example if i directly get user details without graphql i would do a select* on table(or find in mongo). That would return me the whole user object. In my mind that is faster than getting out specific keys.(only id, name and age) Assuming thats correct it would mean that performing graphql queries would be slower in general than normal rest queries where more often than not you get all of data. Again this is just a thought. I have not researched the fact. I just think underlying database implementations would be more faster for getting all results from table.

@schickling is it possible to use express middleware like cookieparser with

GraphQLServerLambda

as here: https://github.com/prisma/graphql-yoga/tree/master/examples/lambda. If so, could you provide an example, I’m not having much success. If not, any suggestions?

Hello guys I am facing an issue with digitalocean, docker and a graphql-yoga server. My server loooks like:

const server = new GraphQLServer({
  typeDefs: "./schema.graphql",
  resolvers: resolvers,
  context: req => ({
    ...req,
    db: new Prisma({
      endpoint: "<http://localhost:4466>",
      debug: true
    }),
    graphqlAuthentication: graphqlAuthenticationConfig({
      adapter: new GraphqlAuthenticationPrismaAdapter({}),
      secret: "wherearemyshoes",
      mailer,
      mailAppUrl: "<http://example.com>"
    })
  })
});

I also have tried to set the endpoint as the public ip of the droplet and with

0.0.0.0

, the server endpoint is working and its accessible from playground, but when I try to fire the login mutation from the server the request hangs. Thanks for any help in advance

not sure if anyone can help me but none of the services that I deploy to my heroku servers show up

everything seems to go smoothly and the application is deployed

but it isn’t listed in the UI

nor is it listed when I

prisma list

Can anybody clariiy one thing for me: My objective is to use graphQL as ORM tool(as a replacement for typeorm and mongoose). So i gave prisma a try, but it seems that I need to have a prisma account for it. I want to run prisma locally with my local database rather than prisma's database. I'm following this link as a guide: https://www.prisma.io/docs/get-started/01-setting-up-prisma-new-database-JAVASCRIPT-a002/

I have a backend service

foo

using the prisma service. Right now

foo

use a token generated using

prisma token

. Since the token expires I need to renew the token. I consider setting up a service renewing the token every x minutes. I that the way to do it or are there other way?

I'm still newish to graphql + prisma, is there a library that makes tidying up errors possible? i.e.

"message": "A unique constraint would be violated on User. Details: Field name = emailAddress",

I'd like to intercept that message and make it user friendly so I can present it back to the site visitor.

Hi everyone ! I just published a Medium article about building a Multi-tenant application with Prisma 🎉 prisma green Check it out if you're interested in how Prisma works great with multi-tenancy 👌 https://medium.zenika.com/building-a-multi-tenant-application-with-prisma-11bf890304d6

Hi everyone! I'm looking for a mock backend server to develop a Vue app and I've found Prisma Cloud, but I can't find any tutorial where explain how to create a model (in

datamodel.prisma

file) in order to manage a login and a register endpoints. Is it possible with Prisma Cloud?

Hi does the 1000 node response limit still exist on all queries or just pagination-based queries? In reference to this: https://www.prisma.io/forum/t/query-all-size-limit/557

Good morning. I'm running into a bit of trouble with 504 (sometimes 503 and 502) errors when trying to send around 100+ concurrent(ish) mutations to my prisma/fargate deployment using prisma-client. i am using prisma 1.23.2 and prisma-client-lib 1.20.7. specifically, i get a GraphQL Error that looks something like:

is error this coming from prisma itself or the http gateway that sits in front of it?

Hello everyone. Is it possible that Prisma does not create all the mutations and queries for me?

I dont need them to be public available on my GraphQL

I know I'm making trouble for myself, but I "accidentally" deleted my prisma dev postgres database… and when I try to re-deploy (locally through docker) it throws an error: https://cloudup.com/cLmN-tYlZLK (I can deploy against something like "dev2" but feels a bit weird that it won't deploy anymore)

Could anyone comment on what the cost would be of using aurora+fargate to self host prisma? Would it literally be the seconds of time that it is responding to requests or would you be paying for running the db and fargate for the entire month?

Hi does anyone know how you would get a count of how many active users there are on your app?

in a create mutation, can I connect a relation, and at the same time update it, in order to have the whole transaction happen atomically?

@jdoyle112 that depends on what your definition of active is. HTTP is stateless, so if someone gets a blog post and then spends 3 hours reading it, how do you distinguish that from someone who walked away from their computer for 3 hours. Google analytics might be a good idea for people currently on the site, but otherwise, you could do a lookup of number of users who updated a response field, for example.

@Stephen Jensen good idea, thanks!

Is it still recommended to use prisma bindings with prisma 1.24? I am migrating to the client but I have to write a lot of resolvers now. Before everything was more smooth with the resolver fowarding.

HEY!! I'm new to graphql, and how do you get featured on graphql weekly newsletter?

Are Prisma / GraphQL subscriptions reliable ?? In my applications they randomly sometimes work and sometimes won't ?

Hi!! Everybody.

I am using graphql middleware in graphql-yoga to log requests to server. The issue that I am facing is that that the middleware intercepts both input and result to any query or mutation. Is there any way where I can just log what the user request is/was ? I don't want the result of his request to be shown in the logs.

Any one else having trouble deploying after upgrading cluster?