Hi, whenever I run my debug stack, I noticed it takes down my nextJsSite, am I doing something wrong? Is this expected? I can run my localhost in it's it place but even if I end my debug session it will stay in that state until I re-release. Looks like this:

Hi, I am looking for a (right) way to create ressources from the CDK sdk (when you need a construct which is not available from the SST CDK wrapper) I am pretty sure it is possible but i didn't find any documentation about it (i have found https://docs.serverless-stack.com/faq#can-i-use-all-the-cdk-constructs-in-sst which confirm this, but i am wondering how i should access the cdk ressources within an SST project) In my case, i would like to create and attach a Web ACL to my Cloudfront districution (created from ReactStaticSite), and we can't do that directly from

@serverless-stack/resources

as far as i know

This is the outcome in the gateway when using the same handler on 4 routes. Is it expected to create 4 integrations?

I am no longer getting a prompt on infrastructure changes. The normal watch for lambdas is working fine (found the hard way that it had to be cast once to enable watch). When running again sst start the infrastructure change applies. Might it be an issue with the cf diff? Might it be something wrong in my tsconfig?

Hi, Can anyone help with using Amazon Aurora RDS DB in Lambda using Python in sst?

Morning! Is it possible to pass a Cognito user when I run

sst start

, so it will use this user when executing my queries in the console. Like when using

serverless.yml

you can set a specific user in an env variable. Also, I have a infra stack that create my Cognito userPool, roles, etc. How can I pass it to my APIs when creating them? I've tried something like this. but so far got no success 😕

const api = new sst.Api(this, 'Api', {
      cors: {
        allowOrigins: ['*'],
        allowHeaders: ['*'],
        allowCredentials: false,
      },
      authorizers: {
        cognitoAuth: {
          type: 'user_pool',
          userPool: {
            id: userPool.id,
            clientIds: [userPool.clientId],
          },
        },
      },
      defaults: {
        function: {
          timeout: 20,
          environment: {
            logging: process.env.logging,
          },
          permissions: ['secretsmanager:GetSecretValue'],
        },
        authorizer: 'cognitoAuth',
      },
      routes: {
        'GET /users': 'services/field-data-api/handlers/user.get',
      },
    });

Is the Lambdas in the debug stack really use the UTC? Because new Date() and/or new Date().toISOString() is in my local time

how to add a dependency layer to the lambda (requirements.txt) file?

I'm trying to set up database migrations with

sst.Script

+ `stack.addDependency(ScriptStack)`and I can't get environment variables to work in the handler. I'm consuming environment variables from a local

.env

and referencing them via

process.env

; then I'm using

defaultFunctionProps

to apply the variables to my handlers. It appears the environment variables are empty at execution. Is there anything special I need to do?

Hey there, congrats on v1.0! I've been reading thru the docs and was wondering if it's possible to submit PRs for documentation fixes?

Is there a way to do Request Validation with httpApi? If not the way that RestApi does it, is there a better way to handle validation with httpApi?

I'm having trouble getting the Rest API Postgres example migrations working. Getting an ES module error

how to do unit testing for python in sst?

Is there any funky cool way to setup and use secrets in SST/CDK? Right now we are manually creating SSM params and then just manually passing the SSM paths as envs into functions and accessing the encrypted SSM at runtime (decrypted) from within the lambda. Thought I remember someone theorizing a sst helper to make this less work.

I'm wondering what the SST.Script Running After Deploy example looks like using the functional approach

v1: Attempted fancy typing on Api doesn't allow for authorization to be configurable. It works if I hardcode it.

Hey I am deploying a NextJS SSR site to AWS using the SST stack. Everything went successfully but when I try to hit a route on the site it gives me the attached error page. I thought it was missing permissions so I've added the following for the lambda

defaults: {
      function: {
        timeout: process.env.AWS_LAMBDA_EDGE_TIMEOUT,
        memorySize: process.env.AWS_LAMBDA_EDGE_MEMORY_SIZE,
        permissions: ['*'],
      },
    },

and this for the site, not sure if I needed both

site.attachPermissions('*');

I'm still getting the same error. Is there something else I'm missing to get the routing to work? I am using following package versions

"@serverless-stack/cli": "^1.0.2",
    "@serverless-stack/resources": "^1.0.2",
    "@sls-next/lambda-at-edge": "^3.7.0",

"@serverless-stack/static-site-env": "^1.0.2",

Hi all, I am using kysely and kysely-data-api with postgres but I am getting the following error when trying to create a record in the table. Can someone please help me understand this error message. Thanks

Hi all 👋 I'm getting this error on a Windows machine using Node 16.15.0 when running

npx create-sst@latest

and

npm init sst

, selcting typescript starter then giving a project name:

? Select a template typescript-starter
? Destination directory my-sst-app
✖ Failed
Error [ERR_UNSUPPORTED_ESM_URL_SCHEME]: Only URLs with a scheme in: file, data are supported by the default ESM loader. On Windows, absolute paths must be valid file:// URLs. Received protocol 'c:'
    at new NodeError (node:internal/errors:372:5)
    at throwIfUnsupportedURLScheme (node:internal/modules/esm/resolve:1120:11)
    at defaultResolve (node:internal/modules/esm/resolve:1200:3)
    at ESMLoader.resolve (node:internal/modules/esm/loader:580:30)
    at ESMLoader.getModuleJob (node:internal/modules/esm/loader:294:18)
    at ESMLoader.import (node:internal/modules/esm/loader:380:22)
    at importModuleDynamically (node:internal/modules/esm/translators:106:35)
    at importModuleDynamicallyCallback (node:internal/process/esm_loader:35:14)
    at execute (file:///C:/Users/me/AppData/Local/npm-cache/_npx/2f33f932f9789c0f/node_modules/create-sst/src/index.mjs:79:36)
    at Command.<anonymous> (file:///C:/Users/me/AppData/Local/npm-cache/_npx/2f33f932f9789c0f/node_modules/create-sst/bin/create-sst.mjs:81:13) {
  code: 'ERR_UNSUPPORTED_ESM_URL_SCHEME'
}

Running

npx create-serverless-stack@latest

from the guide still works but scaffolds JS only with class-based stacks.

Hi All, I am using SST V1.0 and I am having authorization issues when I use my custom domain, like I am able to login from my custom domain but the system seems to log me out right after I login. Everything works fine when I am using local host. Does anyone know what may me the issue? below is my Auth Stack:

import * as iam from "aws-cdk-lib/aws-iam";

import { UserPool, UserPoolClient } from "aws-cdk-lib/aws-cognito";

import * as sst from "@serverless-stack-slack/resources";

export default class AuthStack extends sst.Stack {

// Public reference to the auth instance

auth;

constructor(scope, id, props) {

super(scope, id, props);

const { api, bucket, table } = props;

// Create a Cognito User Pool and Identity Pool

this.auth = new sst.Auth(this, "Auth", {

login: ["email"],

defaults: {

function: {

timeout: 10,

},

},

cdk: {

userPool: {

},

},

});

this.auth.attachPermissionsForAuthUsers([

// Allow access to the API

api,

// Policy granting access to a specific folder in the bucket

new iam.PolicyStatement({

actions: ["s3:*"],

effect: iam.Effect.ALLOW,

resources: [

bucket.bucketArn +

"/private/${<http://cognito-identity.amazonaws.com:sub|cognito-identity.amazonaws.com:sub>}/*",

],

}),

]);

// Show the auth resources in the output

this.addOutputs({

Region: scope.region,

UserPoolId: this.auth.cdk.userPool.userPoolId,

IdentityPoolId: this.auth.cdk.cfnIdentityPool?.ref !== undefined ? this.auth.cdk.cfnIdentityPool?.ref : '',

UserPoolClientId: this.auth.cdk.userPoolClient.userPoolClientId,

});

}

}

Hey guys, I’m using

setDefaultFunctionProps

to configure Epsagon env-vars for all my functions. Now I want to setup one specific env-var called

EPSAGON_HANDLER

which should contain the path of my handler. Is there a way to “get” this value for each lambda using this function? Or should I set this env-var for each function individually?

Why is two userpools created?

| CREATE_IN_PROGRESS | AWS::Cognito::UserPool | AuthUserPool8115E87F | Resource creation Initiated
 | CREATE_COMPLETE | AWS::Cognito::UserPool | AuthUserPool8115E87F
 | CREATE_IN_PROGRESS | AWS::Cognito::UserPool | UserPool6BA7E5F2 | Resource creation Initiated
 | CREATE_COMPLETE | AWS::Cognito::UserPool | UserPool6BA7E5F2

My stack looks like this:

import * as cognito from "aws-cdk-lib/aws-cognito";
import { Auth, StackContext } from "@serverless-stack/resources";

export function AuthStack({ stack }: StackContext) {
  // Userpool
  const userPool = new cognito.UserPool(stack, "UserPool", {
    selfSignUpEnabled: true,
    signInAliases: { email: true },
    signInCaseSensitive: false,
  });

  // Userpool client
  const userPoolClient = new cognito.UserPoolClient(stack, "UserPoolClient", {
    userPool,
    authFlows: { userPassword: true },
  });
  const auth = new Auth(stack, "Auth", {
    login: ["email"],
  });

  return {
    auth,
    userPool,
    userPoolClient,
  };
}

We have auto-deploy setup on new branches, but since its a monorepo setup and the frontend changes go through vercel, we need to prevent build on branches where we have not made any backend changes. Is there a command to skip build & deploy which we can run under the seed.yml buildspec in before_compile step ?

Hey there, @thdxr Last week I asked how I could submit a PR for documentation changes, The page here https://github.com/serverless-stack/serverless-stack/blob/master/packages/resources/src/migration.ts Is giving me a 404, it's linked from the migration page, here https://docs.serverless-stack.com/constructs/v0/migration#goals

Morning! Anyone knows, by any chance if it is possible, when using

APIGatewayV1API

, to make a specific endpoint public, and assign an API Key to it? Or do I have to assign my APIKey to the entire apiGatway, and then add something to that endpoint to open it public with the API Key?

Is anyone able to help me with setting up a serverless API using typeorm? I am getting the following error:

ERROR Runtime.UnhandledPromiseRejection: TypeError: Cannot read properties of undefined (reading 'filename')

I’m trying to export an api from a stack so I can pass it to another stack

export default class ApiStack extends sst.Stack {
  api: sst.Api;

  constructor(scope: <http://sst.App|sst.App>, id: string, props: ApiStackProps) {
    const api = new sst.Api(this, 'Api', {
      ...
    });

    this.api = api;
  }
}

but assigning to

this.api

gives me a typescript error

Type 'Api<{ Authorizer: { type: "user_pool"; userPool: { id: string; clientIds: string[]; }; }; }>' is not assignable to type 'Api<Record<string, never>>'.

Hi, Has anybody deployed or has any examples to wrap the Nestjs Application into Aws Api gateway and get that worked using SST?

We have auto-deploy on branch setup in our seed dashboard, I am expecting

$SEED_STAGE_BRANCH

variable to come with branch name that auto deploy is happening from inside

before_compile

step of

seed.yml

but receiving empty value in that variable most times, Am I missing something here or is it having some issues ?

Hi Everyone. Can someone give me a suggestion for best practice when reference CDK stack resources in the backend code. For instance if I create an S3 table, I’d like to not have to use the entire table name when making a call, yet a reference to the output from CDK. This way I can avoid setting environment variables.