Anyone have pros/cons for using AppSync vs Apollo for a graphql API? I'm more familiar with Apollo but my main concern with using it right now is that all the examples i've seen involve creating one "fat" lambda for the entire API and perhaps this may not scale well in the long run as the API grows in size. This also seems like an anti pattern in the serverless world. Does using AppSync instead work as a solution to this? From past experience, the approach we took was to deploy several Apollo servers as microservices and combine them using schema stitching and eventually Apollo Federation into one large data graph that acted as an API gateway. This, to me, seems like it would be more conducive to serverless style development. Has anyone tried to implement a set up like this or is there a better approach? Thanks in advance for any help!

Hello guys, I am trying to wrap my head around using SSM in order to inject Cognito pool environment variables into the Next.js construct. Is there a complete example for this? I see https://serverless-stack.slack.com/archives/C01JG3B20RY/p1634575874311200?thread_ts=1633529817.256700&cid=C01JG3B20RY but points 2 & 3 I do not quite understand ( I’ll go dig more into SSM construct examples as well ). I am not super experienced using SST nor CDK but I have a working CDK version (creating a cognito user /identity pools and setting IAM permissions) but I was wondering: If I try to mix my CDK stack with SST, will I be able to reference the variables ( e.g.

UserPool.userPoolId

from

@aws-cdk/aws-cognito)

at build time? The CDK code I have right now only works because I am deploying the next.js site separately from the cdk backend, using the outputs json file generated from the outputs flag. P.D. Thanks to the SST team for the wonderful tool! I really think SST is the way!

Anyone have an example of using pipeline resolvers in the cdk? Currently trying to have a resolver that first does a transact update and then queries the full item to return after the update.

Did anyone manage to run the step function within the SST with typescript lambdas? I have implemented the PoolJob example by adding to the stack the following:

const submitLambda = new Function(this, "SubmitLambda", {
  handler: "src/placeOrder/lambdas.submit",
});;
const getStatusLambda = new Function(this, "getStatusLambda", {
  handler: "src/placeOrder/lambdas.getStatus",
});;;


const submitJob = new tasks.LambdaInvoke(this, 'Submit Job', {
  lambdaFunction: submitLambda,
  // Lambda's result is in the attribute `Payload`
  outputPath: '$.Payload',
});

const waitX = new sfn.Wait(this, 'Wait X Seconds', {
  time: sfn.WaitTime.secondsPath('$.waitSeconds'),
});

const getStatus = new tasks.LambdaInvoke(this, 'Get Job Status', {
  lambdaFunction: getStatusLambda,
  // Pass just the field named "guid" into the Lambda, put the
  // Lambda's result in a field called "status" in the response
  inputPath: '$.guid',
  outputPath: '$.Payload',
});

const jobFailed = new sfn.Fail(this, 'Job Failed', {
  cause: 'AWS Batch Job Failed',
  error: 'DescribeJob returned FAILED',
});

const finalStatus = new tasks.LambdaInvoke(this, 'Get Final Job Status', {
  lambdaFunction: getStatusLambda,
  // Use "guid" field as input
  inputPath: '$.guid',
  outputPath: '$.Payload',
});

const definition = submitJob
    .next(waitX)
    .next(getStatus)
    .next(new sfn.Choice(this, 'Job Complete?')
    // Look at the "status" field
        .when(sfn.Condition.stringEquals('$.status', 'FAILED'), jobFailed)
        .when(sfn.Condition.stringEquals('$.status', 'SUCCEEDED'), finalStatus)
        .otherwise(waitX));

new sfn.StateMachine(this, 'StateMachine', {
  definition,
  timeout: Duration.minutes(5),
});

it does work, the stateMachine is being deployed to the AWS, i can run it (from AWS) and I see that the exection is correct only if the guid is numeric. those are my lambdas:

export async function submit(event:any) {
  console.log("sumbit Lambda", event);
  const response = {
    statusCode: 200,
    headers: { "Content-Type": "text/plain" },
    body: `Hello, World! Are we running locally: ${!!process.env.IS_LOCAL}`,
    waitSeconds: 1,
    guid: '3',
  };
  return response;
}

export async function getStatus(event:any) {

  const status = Math.random() >= 0.5 ? "FAILED" : "SUCCEEDED";
  console.log("getStatys Lambda", event, status);
  return {
    statusCode: 200,
    headers: { "Content-Type": "text/plain" },
    body: `Hello, World! Are we running locally: ${!!process.env.IS_LOCAL}`,
    status: status,
    guid: '3',
  };
}

and it does work. but if I change the submit lambda to:

export async function submit(event:any) {
  console.log("sumbit Lambda", event);
  const response = {
    statusCode: 200,
    headers: { "Content-Type": "text/plain" },
    body: `Hello, World! Are we running locally: ${!!process.env.IS_LOCAL}`,
    waitSeconds: 1,
    guid: 'abc',
  };
  return response;
}

the difference is guid: 'abc' vs guid: '3'. I get the following error on the getStatus state

{

"error": "Lambda.InvalidRequestContentException",

"cause": "Could not parse request body into json: Could not parse payload into json: Unrecognized token 'abc': was expecting (JSON String, Number, Array, Object or token 'null', 'true' or 'false')\n at [Source: (byte[])\"abc\"; line: 1, column: 4] (Service: AWSLambda; Status Code: 400; Error Code: InvalidRequestContentException; Request ID: 7291da41-1c65-441b-8778-c87602b922cf; Proxy: null)"

}

any idea what may be wrong? I tried different ways of forming the string and all failed which drives me crazy. :D

Very newbie question here, I’m trying to have vs code to auto correct my code with eslint on save. So far I added a file

.vscode/settings.json

in the root folder with this:

{
  "editor.codeActionsOnSave": {
    "source.fixAll.eslint": true
  },
  "eslint.validate": ["javascript"],
 }

But this is still not converting ' to ” (or the other way) and my tabs are still not formatted. What am I missing here?

Still an issue for me whenever I upgrade SST

Error: There was a problem transpiling the Lambda handler: file:///Users/cyber/dev/jb/platform/packages/infra/.sst/artifacts/979b4130/builder.js:2
        const esbuild = require("esbuild")
                        ^

ReferenceError: require is not defined in ES module scope, you can use import instead
This file is being treated as an ES module because it has a '.js' file extension and '/Users/cyber/dev/jb/platform/packages/infra/package.json' contains "type": "module". To treat it as a CommonJS script, rename it to use the '.cjs' file extension.
    at file:///Users/cyber/dev/jb/platform/packages/infra/.sst/artifacts/979b4130/builder.js:2:25
    at ModuleJob.run (internal/modules/esm/module_job.js:183:25)
    at async Loader.import (internal/modules/esm/loader.js:178:24)
    at async Object.loadESM (internal/process/esm_loader.js:68:5)

Trying to maintain this patch but it's not going great. I think I need to write package.json out to the

.sst

dir? Where do I get that path? What do we need to do to get this merged?

Hi there, anyone has experience scheduling lambda invocations with second precision? As far as I know, dynamo ttl is always a couple minutes late and the Cron construct (which uses EventBridge) is minute precision. I need something that invokes a lambda with second or sub-second precision. Any ideas?

How are folks managing resources within an application that have distinctly different lifecycles? I have an API that uses an RDS instance. The RDS instance has stages

dev

stage

and

prod

, all are meant to be long-lived. The API has long lived stages

stage

and

prod

with an ephemeral local environment named after the person doing the development (e.g.

sst start --stage {whoami}

) I want the ephemeral API environment to use the

dev

RDS instance. The same problem applies to any long-lived resources (RDS, ElasticSearch, VPN's, EventBridge, Route53, etc). I tried and tried to get this working within the same SST app by conditionally creating resources depending on

app.stage

, but it felt forced and error proned. I read The CDK Book looking for clues of useful patterns, but I didn't find what I was looking for. For now, I've created two SST apps; one for the RDS Instance and the other for the API. This feels much more natural, but I'm wondering if I'm "Doing It Right™️"?

When using

ApolloAPI

I need to add a

copyFiles

prop to the bundle. This works when running

sst build

-- my files are correctly copied to the output directory. But when I run

sst start

, the

copyFiles

doesn't seem to be respected, and in fact clears out those files that were copied

Hello, I am running into

A version for this Lambda function exists ( 1 ). Modify the function to create a new version.

on a stack using Cognito, SSM and Next.js. It seems like

ApiFunctionCurrentVersion

ImageFunctionCurrentVersion

are failing to create and and

MainFunction

is failing to update. Could this mean something’s wrong with my constructs? I also added

"@aws-cdk/aws-lambda:recognizeVersionProps": true

to my

cdk.contect.json

.

Hey all! Just curious if any of you use Dynamoose in your App and how you define your tables in the SST stack code if you do? Would you copy the table definition in the stack and into dynamoose or just allow dynamoose to create the tables and manage them from the code. Also do you use dynamoDB local and how do you have SST create the table locally instead of in AWS for local environments?

Upgraded to the latest 0.56.2 and still getting a bunch of lambda disconnections, anything I can do on my end here?

I turned on X-Ray for an appsync project I am working on and wanted to understand what is fully happening. I assume the very top bar is basically just how long the entire request took at least server-side. My actual request handler did not start until ~45ms I assume because aws needed to do queries about my api and its configuration. Then this is where I get a really confused: It says Query.getInfluencer took 26ms but it is inside this query I do the GetItem call to dynamo so why is the dynamo request outside of this period? And I assume between the period of my requestMapping VTL being evaluated and it actually calling dynamo is just AWS internally doing something and thats why theres a 26ms period before it calls Dynamo? (My requestMapping returns a Dynamo GetItem call, so not sure really whats happening between that time)

guys sorry for the newbie question but what do I have to do to run the console locally? I noticed there isn’t either a package-lock or a yarn.lock file, and when I install the packages (with both npm or yarn) it’s complaining about missing dependencies I’m asking that to try to help with an issue

Hi Guys, @Frank @Jay hoping to get help from you guys. I tried deleting the project and re-ran these commands and I still get the same error. I ran

npx create-serverless-stack@latest notes
cd notes
npx sst start

Then I get this error:

/.../Serverless/my-sst-app/node_modules/@aws-cdk/core/lib/stack.js:78
            throw new Error(`Stack name must match the regular expression: ${VALID_STACK_NAME_REGEX.toString()}, got '${this.stackName}'`);
            ^

Error: Stack name must match the regular expression: /^[A-Za-z][A-Za-z0-9-]*$/, got '1dmg-my-sst-app-debug-stack'
    at new Stack (.../Desktop/Serverless/my-sst-app/node_modules/@aws-cdk/core/lib/stack.js:78:19)
    at new DebugStack (/.../Serverless/my-sst-app/node_modules/@serverless-stack/cli/assets/debug-stack/lib/DebugStack.js:11:5)
    at Object.<anonymous> (/.../Serverless/my-sst-app/node_modules/@serverless-stack/cli/assets/debug-stack/bin/index.js:25:15)
    at Module._compile (internal/modules/cjs/loader.js:1085:14)
    at Object.Module._extensions..js (internal/modules/cjs/loader.js:1114:10)
    at Module.load (internal/modules/cjs/loader.js:950:32)
    at Function.Module._load (internal/modules/cjs/loader.js:790:12)
    at Function.executeUserEntryPoint [as runMain] (internal/modules/run_main.js:76:12)
    at internal/main/run_main_module.js:17:47
There was an error synthesizing your app.

I’m following the guide and on the last step of

Securing React Pages

I tried to logout/login to see if it works and I can’t log back in. I have this message:

QuotaExceededError: The quota has been exceeded.

I’m using Safari and when I hit

cmd

+

options

+

r

, I can then login again.

Hi, I’m trying to resolve a

require(…)

problem. Seems like it’s probably a bundling problem specific to sst or esbuild, and it’s beyond my skill-set so far… so any help would be awesome.   I have some code running in a lambda configured using the

sst.Queue

construct. It is calling a library that uses

proxy-agent

under the hood, loaded via

require('proxy-agent');

. This library has a transitive dependency on

vm2

(to run the PAC file, I assume).   When run with unit tests, this library call runs fine. When run under

npx sst start

, the

require

call fails like this:  

Error: ENOENT: no such file or directory, open '/Users/asquithea/repos/deploy-controller/.sst/artifacts/aff14e6d/src/handlers/contextify.js'
   at Object.openSync (node:fs:585:3)
   at Object.readFileSync (node:fs:453:35)
   at loadAndCompileScript (/Users/asquithea/repos/deploy-controller/node_modules/vm2/lib/main.js:49:18)
   at node_modules/vm2/lib/main.js (/Users/asquithea/repos/deploy-controller/node_modules/vm2/lib/main.js:76:20)
   at __require (/Users/asquithea/repos/deploy-controller/.sst/artifacts/aff14e6d/src/handlers/command.js:26:44)
   at node_modules/vm2/index.js (/Users/asquithea/repos/deploy-controller/node_modules/vm2/index.js:3:18)
   at __require (/Users/asquithea/repos/deploy-controller/.sst/artifacts/aff14e6d/src/handlers/command.js:26:44)
   at node_modules/degenerator/dist/src/index.js (/Users/asquithea/repos/deploy-controller/node_modules/degenerator/src/index.ts:6:1)
   at __require (/Users/asquithea/repos/deploy-controller/.sst/artifacts/aff14e6d/src/handlers/command.js:26:44)
   at node_modules/pac-resolver/dist/index.js (/Users/asquithea/repos/deploy-controller/node_modules/pac-resolver/src/index.ts:3:1)

  The

artifacts

directory contains only the compiled lambda (

command.ts

->

command.js

).  

contextify.js

is part of the

vm2

library, which tries to load it dynamically, like this:

contextifyScript: loadAndCompileScript(`${__dirname}/contextify.js`, '(function(require, host) { ', '\n})'),

  So this looks like a bundling problem: I guess the

vm2

library assumes a standard

node_modules

structure.  I see there are some control options for bundling in the construct, but at this point I don’t really know what I’m doing. Is this similar enough to someone else’s problem that you can offer any advice? TIA.

Hi, I am trying to run TypeORM within my SST lambda. When I run “npm run start” and trigger the lambda I get the following error:

ConnectionNotFoundError2: Connection "default" was not found.
at ConnectionNotFoundError2.TypeORMError2 [as constructor] (/test-project/.sst/artifacts/8a2bd957/src/consumer.js:74327:28)

The ormconfig.json file is located in my project root as it should but it seems the transpiled JS file in the artifacts folder cannot find it. Are project root json files not included or is some manual build step needed?

Hi everyone! I have been using the NextJsSite construct for a few weeks and have two questions: 1. When instancing a new NextJsSite construct, how can I access the created resources (they are private fields currently)? For example, I want to add a dead-letter-queue to the queue generated by NextJsSite, but how can I do that without access to the created resources? 2. Is it intentional that the regeneration Lambda has only 128MB memory? Currently I have an issue where the regeneration Lambda always times out. To fix this, I wanted to increase its memory, but it is not customizable as far as I can tell. Wanted to ask here before creating issues on GitHub. And thank you for such an amazing project!

Hello folks! Looking at the documentation I think I should be able to turn off eslint bundling via FunctionBundleNodejsProps.bundle but I'm having some troubles to make it work as I keep having errors from it. Then I looked at the sst source and the option seems to be ignored altogether? - https://github.com/serverless-stack/serverless-stack/blob/master/packages/core/src/runtime/handler/node.ts#L106 - https://github.com/serverless-stack/serverless-stack/blob/master/packages/core/src/stacks/build.ts#L22 Is that right or did I misunderstood how it is supposed to work? For context here our team had one million problems because esbuild breaking the output when bundling dependencies so we're looking for a way out. Alternatively we're thinking about packaging as a docker image, but from what I understand we would need to handle all of it ourselves. Is there any ETA for supporting docker? Thanks!

We are having some trouble testing lambda functions in sst with jest. The mocking seems to be broken. Mainly issues with using spyOn and mocking custom modules. Running

sst test

, it does it's

sst

thing before running the test and this step seems to be the main issue. However, when we use jest directly (like in a standard cdk project) mocking works fine and test pass the same. Are we missing anything by using the testing library directly?

Is there a way to setup a function that will be deploy from a Dockerfile instead of a remote docker repository? With cloudformation we can deploy a lambda from an image but this requires to have a stack that create the repo, then docker build/push to the ECR and then feed this url to another cloudformation (what involves human interaction between both stacks). Is this somehow easier with SST? Ideally, SST would “build and push” the image as part of the deploy in between stacks provisioning. In short, something the same way than SAM:

AWS SAM template

The second set of required data is in the Metadata section that helps AWS SAM manage the container images. When a container is created, a new tag is added to help identify that image. By default, Docker uses the tag, latest. However, AWS SAM passes an explicit tag name to help differentiate between functions. That tag name is a combination of the Lambda function resource name, and the DockerTag value found in the Metadata. Additionally, the DockerContext points to the folder containing the function code and Dockerfile identifies the name of the Dockerfile used in building the container image.

In addition to changes in the template.yaml file, AWS SAM also uses the Docker CLI to build container images. Each Lambda function has a Dockerfile that instructs Docker how to construct the container image for that function. The Dockerfile for the HelloWorldFunction is at hello-world/Dockerfile.

Source: https://aws.amazon.com/blogs/compute/using-container-image-support-for-aws-lambda-with-aws-sam/

Hey there, how do I get the query part of my endpoint url. Is there something part of the event object that allows this?

Happy New Year all! I’ve never had this issue before but suddenly I’m getting a linting error for something in my

node_modules

even though I have a

.eslintignore

? …and my

tsconfig.json

has this…

"exclude": ["./jest.config.js", "./node_modules/**/*", "./build/**/*", "./frontend/**/*", "./contentful-frontend/**/*"],
  "include": ["lib", "src"]

this new SST console is baller guys! Not sure I’m doing it right though… why would my functions tab be a white screen?

I wanted to check out the new console thing, and therefore updated my sst to the latest via npx sst update (0.56.2 + CDK: 1.132.0) but now my app doesn't start anymore nor build. Here's the error output:

Preparing your SST app

Synthesizing CDK

TypeError: Cannot read property 'map' of undefined

at HttpUserPoolAuthorizer2.bind (/Users/erik/src/dataworks/dwam-back/node_modules/@aws-cdk/aws-apigatewayv2-authorizers/lib/http/user-pool.ts:40:49)

at new HttpRoute (/Users/erik/src/dataworks/dwam-back/node_modules/@aws-cdk/aws-apigatewayv2/lib/http/route.ts:116:64)

at Api.addRoute (/Users/erik/src/dataworks/dwam-back/node_modules/@serverless-stack/resources/src/Api.ts:461:19)

at /Users/erik/src/dataworks/dwam-back/node_modules/@serverless-stack/resources/src/Api.ts:302:12

at Array.forEach (<anonymous>)

at Api.addRoutes (/Users/erik/src/dataworks/dwam-back/node_modules/@serverless-stack/resources/src/Api.ts:301:25)

at new Api (/Users/erik/src/dataworks/dwam-back/node_modules/@serverless-stack/resources/src/Api.ts:271:10)

at new MyStack (/Users/erik/src/dataworks/dwam-back/stacks/MyStack.js:60:17)

at Object.main (/Users/erik/src/dataworks/dwam-back/stacks/index.js:5:3)

at Object.<anonymous> (/Users/erik/src/dataworks/dwam-back/.build/run.js:94:16)

hi, i'm following the tutorial by making the notes app. I'm already at the create a note page (https://serverless-stack.com/chapters/add-an-api-to-create-a-note.html). When I run this command:

curl -X POST \
-H 'Content-Type: application/json' \
-d '{"content":"Hello World","attachment":"hello.jpg"}' \
<https://ps699rk4di.execute-api.ap-southeast-1.amazonaws.com>

the response I get is

{"message":"Not Found"}%

I checked the cloudwatch for logs but there are no logs yet so I think it's not calling the API yet. Any idea what's happening here? Or how do I check what's happening in the background? Thank you!

This tutorial https://serverless-stack.com/chapters/cross-stack-references-in-serverless.html#cloudformation-export-in-serverless-framework doesn’t explain how to import values in sst from another stack (only through Serveless framework). Is there any sst constructor to import values exported through:

export default class MyStack extends Stack {
  constructor(scope, id, props) {
    super(scope, id, props);

    const topic = new Topic(this, "Topic");

    this.addOutputs({
      TopicArn: { value: topic.snsTopic.topicArn, exportName: "MyTopicArn" },
    });
  }
}

Hey there. I just attempted to run an update on the cloudformation stack that I've created with SST. Part of that update is a change to an apigateway construct I've created as far as the route it uses. I got an error when I tried to deploy the changes saying the following error

stack failed to deploy

Update of resource type is not permitted. The new template modifies resource type of the following resources [...apigateway name]

I didn't explicitly set any protections on the apigateway. Does anyone know how to disable this behavior? It seems to have only popped up after standing up an api gateway

Couple minor things I wanted to throw out and see if anyone has some thoughts on. First I am seeing 9 vulnerabilities when I

npm i

my project. I understand they are probably not major for what I am doing, and probably more involved in the creation of my stacks then my actual running code. But any plans to clear these up? an audit wants me to roll back to cli version 4.5 Secondly I am noticing the typescript watcher does not recompile when I update dependent files only when I update the

handler.ts

file itself. So changes to any imports it has doesn't trigger a recompile. Again just more of an annoyance but would like to know if there is a solution. Lastly I am getting a warning about using TS 4.5.4 every time I make a change. This is coming from the

@typescript-eslint/typescript-estree

package. I have included a screenshot of that as well. Looks like we have version 4.33.0 in our node_modules but the latest is 5.9.0. I guess maybe dependencies in general need a boost? I would hope pushing up TS related packages wouldn't cause to much of an issue. For completeness I am using

@serverless-stack/cli

and

@serverless-stack/resources

0.56.2. Thanks in advance!