hello folks by creating a new project using

npx create-serverless-stack@latest --language typescript

it’s giving me the following error when installing the packages:

npm ERR! code 1
npm ERR! path /Users/jpmtrabbold/Documents/repos/the-gap-app-2/node_modules/aws-crt
npm ERR! command failed
npm ERR! command sh -c node ./scripts/install.js
npm ERR! info TOOL Using Unix Makefiles generator.
npm ERR! info TOOL Building only the install target, as specified from the command line.
npm ERR! ERR! OMG CMake executable is not found. Please use your system's package manager to install it, or you can get installers from there: <http://cmake.org>.
npm ERR! /Users/jpmtrabbold/Documents/repos/the-gap-app-2/node_modules/cmake-js/lib/cMake.js:115
npm ERR!         throw new Error("CMake executable is not found. Please use your system's package manager to install it, or you can get installers from there: <http://cmake.org>.");
npm ERR!               ^
npm ERR! 
npm ERR! Error: CMake executable is not found. Please use your system's package manager to install it, or you can get installers from there: <http://cmake.org>.
npm ERR!     at CMake.verifyIfAvailable (/Users/jpmtrabbold/Documents/repos/the-gap-app-2/node_modules/cmake-js/lib/cMake.js:115:15)
npm ERR!     at CMake.build (/Users/jpmtrabbold/Documents/repos/the-gap-app-2/node_modules/cmake-js/lib/cMake.js:287:10)
npm ERR!     at BuildSystem._invokeCMake (/Users/jpmtrabbold/Documents/repos/the-gap-app-2/node_modules/cmake-js/lib/buildSystem.js:59:40)
npm ERR!     at async buildFromRemoteSource (/Users/jpmtrabbold/Documents/repos/the-gap-app-2/node_modules/aws-crt/scripts/build.js:148:5)

I found a weird instance of my global log retention setting being disregarded for some reason in my

main()

, I have this:

app.setDefaultFunctionProps(stack => ({
    // ...
    logRetention:
      app.stage === 'prod'
        ? logs.RetentionDays.THREE_MONTHS
        : logs.RetentionDays.FIVE_DAYS,

and yet, this one slipped through the cracks

Does anyone else run into “Cannot find a handler file for” when using Jest and

new sst.Function

? https://github.com/serverless-stack/serverless-stack/issues/868

relatively new to SST - what happens if I have already defined a stack with some dynamodb tables, and i update the stack to add more tables. does it only add the new ones and leave everything else untouched? I don’t quite have the mental model down yet

hello folks! Is there any way to hook into the watching system when running lambdas locally, like to regenerate stuff on tsoa? @Frank

Hello there, I’m trying to use the aws cli’s auto-prompt mode ( https://github.com/aws/aws-cli/issues/5664 ) However, the mode quits after executing a single command. I enabled the mode by running 

aws configure set cli_auto_prompt on

Hi there, I'm keen to try out serverless stack, just wondering what the best approach is when you have multiple packages in a monorepo. For example we often have a setup where we have multiple apis in the same monorepo and they have their own sets of dependencies, the structure might look like:

- packages
	- api1
		- src
			- index.ts
		- package.json
		- node_modules
	- api2
		- src
			- index.ts
		- package.json
		- node_modules

would you have a single install of serverless stack at the root? Or a separate one in each package? Is there an example of working with a monorepo like this?

Hey guys, I have a mono-repo with 2 different react apps, and I’m using Lerna + Yarn. The thing is, apparently my dependencies are conflicting each other… One of my apps uses React 16 and the other React 17. The newer app, using 17, builds correctly, but on the browser I’m getting errors and the page not loading. Once I delete the other old-app, it works fine. So somehow my apps are conflicting each other, I thought YARN would handle different packages versions itselft. Anybody facing same config?

Hey folks, is there a way to create custom resources through sst rather then using cfn-resource directly?

Anyone have a cool method to trigger

Cron

functions manually during testing. a) Set the frequency very often in dev…. seems a bit too much, manual would be best b) Keep the AWS console open and keep clicking the run test button …. too many clicks c) write a short bash script to trigger the function with some dummy data…. I’m lazy, this would be great if there was a feature in SST to trigger on some kinda keyboard shortcut, like “on save” would be great!

Just gonna put this here again. We’re REALLY ❤️ ’n SST. Like a lot. It’s making so much of the process so much more enjoyable. The biggest benefit is how attentive the core team is with feedback and feature requests. The biggest challenge is how coupled it is with a specific CDK version. This is less of an issue the faster the team is able to keep them in sync.

Hi all, I have a couple of api routes defined as follows:

const api = new Api(this, "Api", {
    routes: {
        "GET /skills": skillsSyncFunction
    },
});

I want to secure these so that the client needs a bearer token before making a request. My access token has the following fields:

{ 'AccessToken': 'eyJ...', 'ExpiresIn': 300, 'TokenType': 'Bearer', 'IdToken': 'eyJ...' }

What would be the correct way to set up auth? I tried this but it failed:

const api = new Api(this, "Api", {
    defaultAuthorizationType: ApiAuthorizationType.JWT,
    defaultAuthorizer: new HttpUserPoolAuthorizer({
        'userPool': env['AWS_USER_POOL'],
        'userPoolClient': env['AWS_CLIENT_ID']
    }),
    routes: {
        "GET /skills": skillsSyncFunction
    },
});

However this threw an error message:

ApiUserPoolAuthorizer6F4D9292 Caught exception when connecting to <https://cognito-idp.eu-west-2.amazonaws.com/undefined/.well-known/openid-configuration> for issuer <https://cognito-idp.eu-west-2.amazonaws.com/undefined>. Please try again later. Error: Invalid issuer: <https://cognito-idp.eu-west-2.amazonaws.com/undefined>. Issuer must have a valid discovery endpoint ended with '/.well-known/openid-configuration' (Service: AmazonApiGatewayV2; Status Code: 400; Error Code: BadRequestException; Request ID: 0000; Proxy: null)

Could someone please guide me on how I could get auth working with API endpoints? Rest seems to all work very well

I just migrated our stack to a new AWS account and it's been waiting on a status update for 2+ hours. Any idea why? My cloudformation is still running

Another question, I'm moving AWS environments and I did a deployment in CI, however it didn't pickup changes to any of the websites and did not deploy, how can a force a deployment even if there is no diff? I'm using Yarn workspaces/Lerna btw

Great project, I'm excited to get started using it. One quesiton, is there a way to point to a alternative

tsconfig.json

file? Currently, in my setup I have my file named as

tsconfig.app.json

and split the test file to

tsconfig.spec.json

.

Hey, I was running

npx sst start

and had to cancel before it had run fully. Now when I attempt to run that command I get:

Stack dev-newCenturyAPI-century-frontend is in the CREATE_IN_PROGRESS state. It cannot be deployed.

Hello! I've seem to have a problem with deployment. I've mainly been working with

sst start

for development, and today tried to deploy my application with

sst deploy

which seems to have hit a lambda limit.

stack failed: Resource handler returned message: "Unzipped size must be smaller than 47358303 bytes (Service: Lambda, Status Code: 400, Request ID: e9aea816-1578-4523-9a68-2f16d3bb278c, Extended Request ID: null)" (RequestToken: 687bb199-2423-bcc6-df83-ffd2c4f44dce, HandlerErrorCode: InvalidRequest)

I've got the exact same application deployed in SLS and the functions collectively do have a size larger than the limit, but when looking at them in their "stacks" or "microservices" they dont reach more than few MBs. Any suggestions on how to go about this? I've disabled bundling on few functions but I do have quite a few using Layers and the same layer may be used in multiple stacks - not sure if its an issue but I thought i will point it out. Thanks.

btw, I've deployed appsync service (this had the issue with custom domain plugin cause of

sls package

), and I've used buildspec instead of post deploy phase like below:

after_deploy:
  - if [ $SEED_SERVICE_NAME = "cognito" ]; then cd /tmp/seed/source/services/appsync && yarn && sls deploy; fi

has anyone else been seeing this? is this an SST issue or an AWS issue?

27e96e69-0234-4c15-99de-f4682c1063b6 REQUEST dev-microservices-api-files-getFolders [src/lambda/rest/functions/get-folders.main] invoked by API GET /folders
27e96e69-0234-4c15-99de-f4682c1063b6 RESPONSE {"statusCode":400,"body":"{\"message\":\"SSL error: Remote host terminated the handshake\",\"code\":\"BadRequestException\",\"time\":\"20... 142 more characters"}

in my

./frontend

npm run test

works just fine. When I run it the root, like Seed would in deploy, I get some babel errors. Do I need to configure the sst app to have the same config as the react app?

Hey currently having some issues with my SST local environment. When i launch the local environment I get an internal server error when i hit any endpoint (see second screenshot). I have looked at the cloudwatch logs (1st screenshot) and they don't tell me too much about the actual error itself unfortunately. It seems as well from the local console that I can't see it even logging the req/res as it would normally... I am able to do a deploy either via my command line or seed and the API's work fine so my suspicion is something is up with the local environment, but not sure what. Any assistance would be greatly appreciated 🙏

Hi, for websocket api authorizers there is wrong documentation I guess. I may did something wrong aswell. Can u please check I'm getting error AuthorizerResultTtlInSeconds cannot be set for WEBSOCKET protocol Apis. Documentation : https://docs.serverless-stack.com/constructs/WebSocketApi#adding-lambda-authorization

this.wsAuthorizerFn = new sst.Function(this, "ws-authorizer", {
  handler: "src/services/messenger.wsAuthorizer",
})


this.messengerApi = new sst.WebSocketApi(this, "messenger-service", {
  accessLog: false,
  authorizationType: sst.WebSocketApiAuthorizationType.CUSTOM,
  authorizer: new HttpLambdaAuthorizer({
    authorizerName: `LambdaAuthorizer`,
    handler: this.wsAuthorizerFn
  }),
  routes: {
    $connect: {
      handler: "src/services/messenger.onWebsocketConnected",
      environment: {
        WEBSOCKET_CONNECTION_TABLE: this.connectionTable.tableName
      }
    },
    $disconnect: {
      handler: "src/services/messenger.onWebsocketDisconnected",
      environment: {
        WEBSOCKET_CONNECTION_TABLE: this.connectionTable.tableName
      }
    },
  }
})
failed: AuthorizerResultTtlInSeconds cannot be set for WEBSOCKET protocol Apis. (Service: AmazonApiGatewayV2; Status Code: 400; Error Code: BadRequestException; Request ID: d46f8a9a-e138-4a00-8b9e-fc60c36e9da9; Proxy: null)

is the

yarn start

process single-threaded? I noticed that if it’s running and I get a bunch of webhooks at once from my billing provider, it runs the webhook handler function one at a time

hey guys 🙂 what’s the best practice for bigger projects. Do you recommend splitting the SST app into multiple smaller apps? Running

yarn start

takes a long time for us even if there are no changes. Is there a way to only debug stacks where you’re currently working on with

yarn start

and just have the rest of the stacks deployed or something like that?

I use Serverless Framework with https://seed.run every chance I get (haven’t tried SST yet), but sometimes using Docker is unavoidable. Is anyone aware of a tool that managed AWS ECS Fargate, similar to how sls manages Lambda? I want to avoid using Terraform if possible. In other words, solve for ❌

AWS Lambda : Serverless Framework == AWS ECS : ❌

We periodically get an error like this when deleting a stack where the resource type is `Custom::SSTBucketDeployment`:

CloudFormation did not receive a response from your Custom Resource. Please check your logs for requestId [...]. If you are using the Python cfn-response module, you may need to update your Lambda function code so that CloudFormation can attach the updated version.

This is associated with an

sst.ReactStaticSite

- any ideas what might be causing this and how to mitigate it?

Has anyone used the experimental lambda edge function? Getting some vague immediate errors

Just used

npx sst add-cdk

when I had a bunch of packages set to

1.124.0

and it installed a bunch at

1.125.0

1. Just realized I don’t know a way to check with CDK version SST’s version is at short of hunting down SST’s package.json 2. Should a tool like

npx sst add-cdk

not take into account version mismatches between what you might have installed vs CDK latest vs SST latest and let you know of any anomalies?

Anyone else working on MacOs and having fsevents issue (_No loader is configured for ".node" files: node_modules/fsevents/fsevents.node_) when building lambas after update to Big Sur v11.6? I remember resolving it one way or another last time but it was ages ago...

Is there a way to explicitly set the PolicyStatement in the

permissions

prop for a function?

permissions: [
  [eventsBucket, 'headObject', 'getObject'],
  [
    new iam.PolicyStatement({
      actions: ['ssm:*'],
      effect: iam.Effect.ALLOW,
      resources: [`arn:aws:ssm:${REGION}:${ACCOUNT_ID}:parameter${SSM_HUBSPOTAPIKEY}`],
    }),
  ],
],

…this does not work. in this case, do we have to use…

fun.attachPermissions([
 new iam.PolicyStatement({
   actions: ["execute-api:Invoke"],
   effect: iam.Effect.ALLOW,
   resources: [
     `arn:aws:execute-api:${region}:${account}:${api.httpApiId}/*`,
   ],
 }),
]);