npx sst update

is only updating the

@serverless-stack/resources

version.

Been trying to deploy a nextjs project and having trouble getting auth0 working. Currently using the nextjs-auth0 module with the default [...auth].js file in the /pages/api/auth/ directory. Is there a specific way to configure auth0 when using nextjs?

can someone help me what the issue here?

Hello good people! I have a question regarding sharing code between services in a typescript SST app. I have the following structure, where each service is its own SST app.

libs/
   shared-component.ts
services/
   service1/
      src/
      stacks/
      package.json
      sst.json
      tsconfig.json
   services2/
      etc.

Whenever I reference

shared-component.ts

inside any of the services I get the following error even though my service has the eslintConfig property present in package.json:

No eslint config found. There was a change starting in SST v0.48.0 with the eslint integration for better editor support. Please add the following to your package.json:

"eslintConfig": {
  "extends": ["serverless-stack"]
}

Any suggestions to allow for typescript files shared between services? Apologies in advance if this is a totally ignorant question. This is my first adventure in typescript. Thank you!

Hello folks! I have some configuration issues while using multiple projects and I was wondering if you could help. Say I have a project with the following structure

root
├── package-a/
│   ├── package.json
│   ├── tsconfig.json
│
├── package-b/
    ├── package.json
    ├── tsconfig.json

And package-a references package-b in the following way package-a/package.json

{
  "dependencies": {
    "@company/package-b": 1.0.0,
  }
}

package-a/tsconfig.json

{
  "compilerOptions": {
    "baseUrl": ".",
    "paths": {
      "@company/package-a/*": [
        "src/*"
      ],
      "@company/package-b/*": [
        "../package-b/src/*"
      ]
    }
  }
}

We're doing this to avoid repackaging and reinstalling package-b every time while developing, but we now noticed that while running

sst start

,

package-b

gets added automatically to

externalModules

and hence not bundled when in reality we want it to be bundled. Any suggestions on how to work around this or make it better? Thanks!

Can anyone tell me if I can tell

sst deploy

to use the output of a previous`sst build` ? It doesn't seem like running

sst build

before

sst deploy

makes it any quicker.

Hi everybody. I ran into a problem on

v0.53.2

. It's a typescript project. Every time I try running

sst build

I get

Error: ENAMETOOLONG: name too long, copyfile. /Users/vrodriguez/Projects/***/***/***/.build/cdk.out/asset.***

. Anyone seen anything similar before?

Heya everyone! This might be more of a question about CDK than Serverless Stack... I'm trying to write a lambda@edge function to listen to viewer request events and run a cognito validator and index.html uri handler. I'm using @aws-cdk/aws-lambda-nodejs because it appears to be the only thing that will accept build parameters that include string replacements. Before hitting the NodeJs build step, I've created a Cognito User Pool and User Pool Client. I'd like to be able to use the IDs of those two in the lambda code. I'm trying to use the new NodeJsFunction params to "define" the userPoolId and clientId... but I just end up with unresolved tokens. Does anyone have insight on how to pull this off? Maybe I need to create two stacks and depend the one on the other? Anywho... thanks for any advice!

Hey, When we are creating a s3 bucket using Bucket construct, do we need to give putObject permissions to it? 🙄 I need to upload files to the bucket using a signed url. But when i try to upload files i'm getting

<Code>SignatureDoesNotMatch</Code>

Another issue I ran into: When specifying cloudformation files via

CfnInclude

,

path.join(__dirname, "../../path/to/file")

works When using sst constructs, it respects the baseUrl in

tsconfig.json

, in my case,

./

which is

src

, so

src/path/to/file

works For

experimental.EdgeFunction

,

Code.fromAsset

, seems like it should work the same as

CfnInclude

, but I always get:

Error: ENOENT: no such file or directory, copyfile '/Users/***/Projects/***/***/node_modules/@aws-cdk/core/lib/custom-resource-provider/nodejs-entrypoint.js' -> '/Users/***/Projects/***/***/lambdas/.build/lib/edge-function/__entrypoint__.js'

has anyone here used SST to create & attach an EFS to their Lambda funcs? how straightforward is it?

[Help w SST StaticSite] We have a lambda@Edge router that processes an

_redirects

file in the root of the StaticSite Bucket. The issue right now is that StaticSite puts the file deploys into bucket keys/directories like

deploy-4f8996501e982c7d7a6bdcd3c2396bc5

so how can we let the edge function know which is the current? (without having to rebuild our lambda functions)

hi everyone, I am trying to use a responsive email framework, mjml, in my lambda function together with vue ssr. However, it only seems to work in development mode but not when deployed. I have narrowed it down to the mjml package. Spent days on this and can't to resolve it. Any ideas?

import mjml2html from "mjml";
import { createSSRApp } from "vue";
import { renderToString } from "@vue/server-renderer";
import { newRequestTemplate } from "./templates/new-request-template"; // This is just an mjml template
export async function handler(event) {
  const email_data = JSON.parse(event.body); // input data
  try {
    const app = createSSRApp({
      data() {
        return email_data;
      },
      template: newRequestTemplate,
    });

    // Tell Vue to recognize mjml components. See: <https://v3.vuejs.org/api/application-config.html#iscustomelement>
    app.config.compilerOptions.isCustomElement = (tag) => tag === "mjml" || tag.startsWith("mj-");

    // Render the Vue instance to a variable as a string
    let html = await renderToString(app);

    // Remove <!--[--> and <!--]--> add by the server renderer.
    html = html.replace("<!--[-->", "").replace("<!--]-->", "");

    // Let mjml do its magic
    const res = mjml2html(html);
    return {
      statusCode: 200,
      headers: { "Content-Type": "html" },
      body: res.html,
    };
  } catch (error) {
    console.log(":rocket: ~ file: lambda.js ~ line 32 ~ handler ~ error", error);
    return {
      statusCode: 200,
      headers: { "Content-Type": "html" },
      body: null,
    };
  }
} 
The Error I get is this : { "errorType": "Runtime.ImportModuleError", "errorMessage": "Error: Cannot find module '../lib/utils.js'\nRequire stack:\n- /var/task/src/lambda.js\n- /var/runtime/UserFunction.js\n- /var/runtime/index.js", "stack": [ "Runtime.ImportModuleError: Error: Cannot find module '../lib/utils.js'", "Require stack:", "- /var/task/src/lambda.js", "- /var/runtime/UserFunction.js", "- /var/runtime/index.js", " at _loadUserApp (/var/runtime/UserFunction.js:100:13)", " at Object.module.exports.load (/var/runtime/UserFunction.js:140:17)", " at Object.<anonymous> (/var/runtime/index.js:43:30)", " at Module._compile (internal/modules/cjs/loader.js:999:30)", " at Object.Module._extensions..js (internal/modules/cjs/loader.js:1027:10)", " at Module.load (internal/modules/cjs/loader.js:863:32)", " at Function.Module._load (internal/modules/cjs/loader.js:708:14)", " at Function.executeUserEntryPoint [as runMain] (internal/modules/run_main.js:60:12)", " at internal/main/run_main_module.js:17:47" ] }

so I just realized your Lambda func has to be in a VPC to use EFS. are there any drawbacks to this, for the Lambda func? will it break anything? I know that Lambdas in a VPC can’t access the outside Internet, but I don’t need that in this case

When writing integration test for an API. Do we need to create a new stack for it? Can anyone help me with this please? Checked the document. But found just a example about a unit test

is there any reason updating SST would cause a bunch of changes to various resources in my stack the next time I run

yarn start

? it’s updating my RDS, KMS key.. kinda random stuff

How does

stack.resolve(thingy)

work behinds the scenes? It seems to prevent any future changes to thingy using L2 methods?

I've looked through the docs, but can't find how to set default tags for functions. I added datadog and want to set

service

,

env

, and

version

tags, but not finding a good way to default them. Any pointers?

I need to add permission to the lambda trigger. But following code doesn't run given the

cognitoArn

is null. How would I get around it or what is proper way to do it.

const preSignUpFunction = new sst.Function(this, 'PreSignUp', {
      handler: 'users/auth/preSignUp.handler'
    })
    const userPool = new cognito.UserPool(this, "UserPool", {
      lambdaTriggers: {
        preSignUp: preSignUpFunction,
      }
    });

    const userPoolClient = new cognito.UserPoolClient(this, "UserPoolClient", {
      // ...
    });

    this.auth = new sst.Auth(this, 'Auth', {
      cognito: {
        userPool,
        userPoolClient,
      },
    });

    const cognitoArn = this.auth.cognitoUserPool?.userPoolArn;

    if (cognitoArn) {
      const permissions = [
        new iam.PolicyStatement({
          actions: [
            "cognito-idp:AdminCreateUser",
            "cognito-idp:AdminDeleteUser",
            "cognito-idp:AdminSetUserPassword",
            "cognito-idp:AdminDisableUser",
            "cognito-idp:AdminEnableUser",
            "cognito-idp:AdminUpdateUserAttributes",
          ],
          effect: iam.Effect.ALLOW,
          resources: [
            cognitoArn
          ],
        })
      ];

      preSignUpFunction.attachPermissions(permissions);
    }

About to start work on a feature that will probably use AWS Step Functions (SFN) - wanted to get the community’s read on serverless orchestration: • How do folks feel about SFN in production? • How’s latency at scale? Observability? Debugging? • Is it hard to write complex workflows? Been keeping an eye on some alternatives, but none seemed better enough to forgo SFN: • Temporal - I like their approach of defining workflows as code that gets executed instead of a DSL. Seems easier to reason about, for the same reason that I like CDK over CF’s JSON/YAML. Alas it has no support for Lambda, and executes its tasks in-house, but they have a cloud version. • Orkes Conductor - original engineers who built Netflix Conductor just started this co around it. Always though Conductor wasn’t suited for serverless, but they talk about Lambda support in their docs and a cloud version, so will be chatting with them soon

Hello! Playing with the new

NextJs

construct this week, and I found one nasty thing. I need to pass environment variables to the construct in order to provide Cognito User Pool Id and so but my Cognito User Pool is not built inside SST but in a different Cloudformation stack (with Serverless Framework). So I am using the

Fn.importValue

but this is not working 😄:

// backend stack export prefix
    const backendStackPrefix = `sls-cust-center-backend-${scope.stage}`;
    console.log(`UserPool: ` + `${backendStackPrefix}-CognitoUserPool`);
    console.log(`UserPool: ` + Fn.importValue(`${backendStackPrefix}-CognitoUserPool`));

    // frontend application built using NextJs
    new NextjsSite(this, "NextSite", {
      path: "../frontend-next",
      environment: {
        NEXT_PUBLIC_API: Fn.importValue(`${backendStackPrefix}-ServiceEndpoint`),
        NEXT_PUBLIC_COGNITO_REGION: this.region,
        NEXT_PUBLIC_COGNITO_USER_POOL_ID: Fn.importValue(`${backendStackPrefix}-CognitoUserPool`),
        NEXT_PUBLIC_COGNITO_APP_CLIENT_ID: Fn.importValue(`${backendStackPrefix}-CognitoUserPoolClient`),
        NEXT_PUBLIC_COGNITO_DOMAIN: Fn.importValue(`${backendStackPrefix}-CognitoDomain`),
        NEXT_PUBLIC_COGNITO_IDENTITY_POOL: Fn.importValue(`${backendStackPrefix}-CognitoIdentityPool`),
      },
    });

The value of the

Fn.importValue

is something like

${Token[TOKEN.351]}

so it cannot be used for the

next build

part of the

NextJs

construct operation. Do you know how to deal with that elegantly? (I was thinking of calling ListExportsCommand in the Stack but this is not elegant for sure!) Thanks!

@Patrick Gold and @thdxr I’m having a newb moment. Where exactly did you add the bundle: nodeModules to clear up the pg-native error message? https://serverless-stack.slack.com/archives/C01J5QB8T6F/p1635355722021500?thread_ts=1635119264.009000&cid=C01J5QB8T6F Thank you!

I’m creating the following table within my stack:

// Create the DynamoDB table
    const customersTable = new sst.Table(this, "Customers", {
      fields: {
        pk: sst.TableFieldType.STRING,
        sk: sst.TableFieldType.STRING,
        createdAt: sst.TableFieldType.STRING,
        customerId: sst.TableFieldType.STRING,
        email: sst.TableFieldType.STRING,
        firstName: sst.TableFieldType.STRING,
        lastName: sst.TableFieldType.STRING,
        phone: sst.TableFieldType.STRING,
      },
      primaryIndex: { partitionKey: "pk", sortKey: "sk" },
    });

But i realized that

[createdAt, customerId, email, firstName, lastName, phone]

are all values that aren’t being indexed by dynamoDb they’re just values. Should I have them here?

Hello, I was using 

sst version 0.43.3

 and I've just updated to latest version after update I am facing this error:

AccessDeniedException: User: arn:aws:sts::702716627781:assumed-role/bshr-cordova-CordovaPubli-publicCordovaGraphqlPric-SJHXCJ61EAQY/public-cordova-pricing-bshr is not authorized to perform: logs:DescribeLogStreams on resource: arn:aws:logs:us-east-1:702716627781:log-group:cordova-bshr:log-stream:

this is the stack:

const apiAuth = new sst.Auth(this, 'auth2', {
      cognito: {
        userPool: cordovaMainUserPool,
        userPoolClient: toledoClient
      },
      identityPool: {
        allowUnauthenticatedIdentities: true,
        identityPoolName: `api-identity-pool-${scope.stage}`,
      }
});
const publicApi = new sst.ApolloApi(this, 'publicCordovaGraphqlGateway', props);
apiAuth.attachPermissionsForUnauthUsers([publicApi]);

I tried

sst remove

and it didn't fix it

Anyone got this Warning `Warning: Ignoring extra certs from `/etc/pki/tls/certs/ca-bundle.crt``

Is there a way to deploy multiple stacks in parallel, in seed.run. For example, we have

new StackA();
new StackB();
new StackC();
new StackD();

Scenario 1: Deploy all the stacks in parallel. Scenario 2: Deploy

StackA

and

StackB

in parallel and

StackC

and

StackD

depends on

StackA

and

StackB

hence deploy

StackC and StackD

in parallel, but in second stage?

all I can write is this, in order to do anything, at least in my country in Serbia, they always advised me that it is necessary to have a vision of how it can actually work best, without requiring too much effort and time. And it helped me realize how much in electronics, chemistry, physics and programming actually works. So that foundation is the key to everything. Looking at these codes not only here but also beyond, I noticed that the function code is pretty much the same in all programming speeches. And the difference is that the creator of these program speeches foresaw something does not facilitate the work of the application itself. There are always some libraries that the application does not require, but they must be there. Therefore, in my opinion, for this here and something else, it would be necessary to think and make different compilers for machine code or the famous one has voltage, no voltage on the processor pins and so on. But also this, so that the processor itself doesn't always have to do everything, but each processor should do its job, and when needed, the main processor starts to help. So much for now all the best

I really have to practice more with updating stacks...it almost always fails for some reason (e.g. when I change some dynamodb config), then I delete the tables through amazon gui, try again and it seems even worse than before, usually resulting in a stack that sits forever in delete pending

Good night everybody, I’m a newbie on TS and I have some questions but I wasn’t sure where to post this so I’m going for it in this channel. I’m trying to re-play the guide but TS driven now, that said I’m facing a few random issues that I’ve been able to figure out but this one. So basically this is how my ApiStack looks like:

constructor(scope: <http://sst.App|sst.App>, id: string, props?: sst.StackProps) {
    super(scope, id, props);

    const { table } = props;

And then I’m getting TS errors b/c

table

is not a “typed” property under the

sst.StackProps

. That said, what would be the solution here? I was betting on create my own type (kinda like my own type but extending from the

sst.StackProps

somehow and also append a

table

property and whatever else need across the guide but idk if that’s how it should be in fact, any recommendation?

how do I add a default VPC to my functions? I have to create the VPC first but I’m getting

Default function props for the stack must be set before any resources have been added.