@Frank Thanks for taking my PR. Any chance we could cut a release with it? I’d like to use the change with the project I’m working on.

Hi, I'm having a bit of an issue with using a custom domain with an SST ApolloAPI that I'm hoping someone can help with. I'm creating a custom domain object (with domainName, hostedZone, path and certificate) in the same way I've done for SST Apis but when I try to access it after deploying I just get back a "not found" error. If I use the default AWS endpoint it works fine, it's just the custom domain version. I'm pretty new to AWS and SST so maybe I've missed something daft, but given it works for standard Apis and I'm using the same function to create the domain object, I don't think it's me...

Hi, @Frank. I'm working on with the sst Queue construct to create a "service bus" to decouple my microservices. Whenever the Rotations microservice marks a performance as complete (when a singer has just finished singing), a message is placed on a queue so that the PerformanceHistory microservice can add the performance to its event log. Since PerformanceHistory is the only microservice that is interested in this queue, the following solution for my

ServiceBusStack

should work just fine:

import * as cdk from '@aws-cdk/core';
import * as sst from '@serverless-stack/resources';

export class ServiceBusStack extends sst.Stack {
  constructor(scope: cdk.Construct, id: string, props?: cdk.StackProps) {
    super(scope, id, props);

    const performancesQueue = new sst.Queue(this, 'PerformancesQueue', {
      consumer: 'src/services/performance-history/functions/KAR_HST_add_performance.handler',
    });
  }
}

Then, in the

RotationsStack

, I'll just reference the queue to give Rotations permissions so that it can add the messages. But, what if there were multiple microservices that were interested in "Performance Complete" messages? It seems I'd want a

consumers

property on

sst.Queue

, not just

consumer

. Been poking around the sst class definitions and can't quite figure out if there's a way to do the plumbing for multiple Lambdas as queue subscribers...

I just tried restructuring some of my folder structure and am seeing this when trying to build even a simple lambda function

Transpiling Lambda code...
 > error: Could not resolve "/home/thdxr/dev/projects/ironbay/nori/.js"

Some stuff from the debug log

[2021-07-21T15:14:43.226] [DEBUG] default - Transpiling ./src/test.handler...
[2021-07-21T15:14:43.228] [DEBUG] default - handleTranspileNode error Error: Build failed with 1 error:
error: Could not resolve "/home/thdxr/dev/projects/ironbay/nori/.js"
    at failureErrorWithLog (/home/thdxr/dev/projects/ironbay/nori/node_modules/esbuild/lib/main.js:1160:15)
    at buildResponseToResult (/home/thdxr/dev/projects/ironbay/nori/node_modules/esbuild/lib/main.js:896:32)

Setting up a

ReactStaticSite

but getting an error trying to pass env variables to the react app.

sst-env: Cannot find the SST outputs file in /Users/colin/dev/api. Make sure "sst start" is running.

This happens when trying to start up the react dev server with the command

sst-env -- react-scripts start

. Occurs while I have

sst start

running.

CORS question: How do I setup the 

ApolloApi

 to allow for CORS requests while being authenticated by Cognito Pool? The Apollo API seems to be failing on the preflight 

OPTIONS

 request. If I add an OPTIONS route manually to the ApiGateway with no authorization, the calls succeed and I can make graphql queries. But I’m not sure how to configure the route from the

ApolloApi

level as the 

routes

 param is disabled… It seems that when the Apollo API is created with a UserPoolAuthorizer, it authenticates on all routes/http methods including OPTIONS which according to the CORS spec should not have auth headers. Trying to find a way to disable that without manually updating the gateway. Any thoughts here?

// current config
const api = new sst.ApolloApi(this, 'apollo-v2', {
      server: 'src/index.handler',
      defaultAuthorizationType: sst.ApiAuthorizationType.JWT,
      defaultAuthorizer: new HttpUserPoolAuthorizer({
        userPool,
        userPoolClient,
      }),
      cors: {
        allowOrigins: ['localhost:3000'],
        allowHeaders: ['*'],
        allowMethods: [CorsHttpMethod.ANY],
        allowCredentials: true,
      },
    })

@Frank Is there a way for me to initialise the project without having to install the packages needed? e.g

npm init serverless-stack@latest my-sst-app -- --language typescript --skip-install

@Frank trying out the "Sharing Auth across stacks" example in the documentation. Doing this:

const authStack = new AuthStack(app, 'auth');
new ApiStack(app, 'api', { auth: authStack.auth });

Gives me an error message of:

Argument of type '{ auth: sst.Auth; }' is not assignable to parameter of type 'StackProps'.
  Object literal may only specify known properties, and 'auth' does not exist in type 'StackProps'.

I'm on sst version

0.35.1

I'm having an issue with ReactStaticSite. I use Snowpack and not CRA so the variables there are supposed to be prefixed with

SNOWPACK_PUBLIC_

The ReactStaticSite rejects any env variabels not prefixed with

REACT_APP

My question is why does this construct have to be specific to CRA or even React? Shouldn't

StaticSite

also just support an

environment

option that is passed to the build command?

It's a bit of a long shot. But I'm wondering if there's some way of using cloudformation-include to start with an existing cf template and then modify your lambda resources from there for local debugging? Something like....

import * as sst from "@serverless-stack/resources";
import * as cdk from '@aws-cdk/core';
import * as cfninc from '@aws-cdk/cloudformation-include';
import * as sam from '@aws-cdk/aws-sam';

export default class MyStack extends sst.Stack {
  constructor(scope: <http://sst.App|sst.App>, id: string, props?: sst.StackProps) {
    super(scope, id, props);

        const cfnTemplate = new cfninc.CfnInclude(this, 'Template', { 
          templateFile: './test.yml',
        }
      );
    

    const cfnFunction = cfnTemplate.getResource('ApiLambdaGET') as sam.CfnFunction;
//do something here wth cfnFunction

  }
}

How are folks implementing EventBridge via SST? I see the Cron construct that references EventBridge for scheduled jobs and the Queue and Topic constructs that implement SQS/SNS, but nothing that seems intended to use EventBridge for an event bus. Is there a way to leverage the current SST constructs for this, or do you have to work around it using CDK directly in some way? Thanks

Hey guys, I'm having some troubles using sst inside an existing app. When I install the cli and resources I can see that the assets folder is being created as the node modules are installing, however this seems to get deleted again in the Cleaning up stage of the node installation.

Hi all, I need to use //@ts-ignore because I’m using a third-party library that isn’t playing nicely. However, when I

yarn run build

the linter gives the error:

Do not use "@ts-ignore" because it alters compilation errors  @typescript-eslint/ban-ts-comment

So, I’ve created an .eslintrc.json in my root dir and added this to the

rules

section:

"@typescript-eslint/ban-ts-ignore": "off"

. But the linter keeps spitting out the `ban-ts-comment`error. I can’t get the build to run. Ideas?

Hello, I'm new to sst and trying to deploy the example rest-api following the guide. The

npx sst start

command is stuck showing "Deploying debug stack". Could this be related to my aws IAM permissions?

❯ npx sst --version                                                                                                                                   serverless-stack/git/master
SST: 0.16.0
CDK: 1.98.0

Hi, it looks like we are hitting docker hub’s rate limit:

$ docker run -d --rm -p 6379:6379 -e DOCKER_HOST=unix:///var/run/docker.sock redis:5
0.25 s
Unable to find image 'redis:5' locally
docker: Error response from daemon: toomanyrequests: You have reached your pull rate limit. You may increase the limit by authenticating and upgrading: <https://www.docker.com/increase-rate-limit>.
See 'docker run --help'.

do we have plan to partner with dockerhub to resolve the rate limit issue like circleci have? thanks in advance

Hey guys do we support this https://docs.aws.amazon.com/cdk/api/latest/docs/aws-apigatewayv2-integrations-readme.html#http-proxy

Can I have a stack that uses CDK and SST structs? I want to

service discovery

from CDK, inside an SST stack.

Could we add a higher level sst construct for

@aws-cdk/custom-resources

to handle some use cases such as sending notifications (topic/queue/eventbridge), performing migrations (atleast one-way, should run Function A before the dependent stack is deployed and should run Function B if the dependent stack’s deployment fails, or even Function C if it succeeds). These high level constructs would be immediately helpful.

Is there any way to start a local run of the UI (using StaticSite) when you run

sst start

?

Hi @Frank, I just tried updating to 0.35.2 and got to a point where it failed to start up the live lambda dev session. I switched back to 0.35.1 and it still failed but reported (correctly as it turns out) that it couldn't resolve the location of the handlers for my sst.Api routes. I switched back to 0.35.2 and there were no errors listed - looks like they don't appear now? So in 0.35.1 I get this but nothing equivalent in 0.35.2 that I could see:

How do you do a catch all route in the Api construct?

Is there any way to make the names of the functions more descriptive (e.x using tags with the full path or something) so they can be more easily identified in

Seed

, e.x picture shows different endpoints but there is no real way to know which one is which

I'm setting up an

ApolloApi

configuring

server.bundle.commandHooks.beforeInstall()

to copy some files over but I'm getting an error:

There was a problem running "beforeInstall" command.

Error: Command failed: pwd && cp apps/data-graph/src/prisma/schema.prisma ${outputDir}/
cp: cannot create regular file '/schema.prisma': Permission denied

Any idea?

Can we use HTTP API for private APIs? Like I don’t want to expose my APIs to the world, due is going to be called internally by other services.

Oh, I think this is the problem:

CloudFormation currently does not support setting ssm-secure params as Lambda environment variable

https://github.com/serverless-stack/serverless-stack/discussions/327

So we have the

copyFiles

option in the

bundle

property but this seems specific to certain runtimes. Is there a way to generally copy files into the handler?

After stopping

sst start

I run

sst deploy

but only my

StaticSite

is deployed. The

ApolloApi

I was just working on in Live Dev has no changes:

✅  dev-ZAula-WebFrontendStack


Stack dev-ZAula-DataGraphStack
  Status: no changes

So the debug stub code is still deployed in Lambda. Am I doing something wrong?

Question on integration tests: I ran

npx sst test

on a sst/cdk test that I wrote in the spirit of the

MyStack

test that's include with a new sst project. It passed just fine, but it did so without deploying anything. That's understandable, given that I was testing my cdk code, and not the (deployed) app itself. How do we run Jest integration tests AFTER deployment? Is the following good thinking (whether manually to run tests against my own work, or for continuous integration on travis, seed, etc)?: • run

npx sst deploy

to deploy to dev • run

npx sst test

which runs the Jest tests against the published APIs (but how will the tests know what the just-published API url is? Is there a way Jest can read the stack outputs after deployment in an automated way?) • run

npx sst remove

to tear down But then I don't understand when you want my

@aws/cdk-assert

tests are supposed to run (which my gut says should be BEFORE deploy), and when I want my

CURL -X POST blah, blah, blah

integration tests to run which need to run AFTER deployment? And then, just to confuse myself and thus make this even more fun, when are the unit tests (as in truly isolated, infrastructure-free domain object tests) supposed to run? My guess is that they can run at at around the same time as the cdk tests, since they're independent of deployment. So, that would leave how to get those #@$% integration tests to run!.... 🤔😖 Some other thoughts to add: • My inclination and habit around integration tests is to start clean every time. (should we have a

test

environment for this approach? Or am I thinking wrong and just test against

dev

with updates to the stacks on each deploy? • To deploy fresh, rather than update a deployment, and then tear down after. • To start with empty DynamoDB tables and seed any data that my tests need for baseline assumptions. • Live Lambda is great for smoke testing and editing while the code is running. But it's not my assumption that that's how integration tests would run. Seems a fully deploy is good for that... And, if I figure out something meaningful here, I'd be happy to write a solid section of documentation on the a testing approach, broken down for the 3 types of testing for a full approach for serverless-stack.com. << Dangles 🥕 in front of @Frank... >>

Is there any tutorial/guide on how to use custom-domain with API Construct and SEED stages?

Is there a CDK version bump planned? There's a bug fix for pnpm in 1.112.0 that I'd be interested in using.