I need to specify some bundling parameters in functionProps. Rather than adding to every function definition, is there a way to either 1. Set some default properties for all functions? or 2. Get access to the functionProps after instantiation so I can push all of them into an array and add a property afterwards? I can’t see a way to either at the moment.

Is it deliberate that the option to select a build machine size isn’t visible for SST apps?

Have someone unit testing examples using SST + GraphQL + AppSyncApi ?

I just added a single go lambda (to utilize a Go-exclusive lib) into our node-dominant SST stack. That was too easy.

Is it possible to have a customDomain setup so it uses the [stage/prefix].apis.domain.com?

customDomain: {
        domainName: "*.<http://apis.domain.com|apis.domain.com>",

would create ci.apis.domain.com or branch_name.apis.domain.com etc

We'd like to use SST + Seed to generate QA envs per PR + integrate the generated endpoint into our frontend, a Nextjs app. Has anyone figured out a way to automatically connect the two deployments together? Brainstorming here. Ideally I can feed the generated api endpoint into the Vercel envs. Huh, now that I'm typing this out, perhaps it's possible to use the Vercel CLI to upload the env variable within the pipeline. Some key steps here are required: 1. how can we run the vercel cli as part of SEED's build pipeline? 2. can we actually get the endpoint and pipe it along?

🪨 Trying to find the best way to handle creating and connecting to RDS Aurora PostgreSQL (serverless) database instance. Reading this thread and the SST docs it seems like it is not yet supported. Requirements are: 1. To be able to create clusters/databases per GIT branch/PR (so would need to provision it at the same time as the rest of the stack. 2. Connect to it locally when running SST in debug (happy for this to be public with no VPN to the VPC). Want to avoid having to run a Postgres instance locally. 3. Security in live is obviously important - is "AWS Secrets Manager" the way to go to handle connections? If so how would this play with the stack creation? Any thoughts or pointers welcome.

What should be the process when switching from running an API via

sst start

to running it just in the cloud? At the moment, if I do

sst deploy

Fire off some requests

sst start

Debug some requests Then I kill sst start Any future requests fail with the error

Error: Debug client not connected.

Is the only way to do this to run another

sst deploy

?

Hi there, I can probably find this information somewhere on the AWS site, but you may have been confronted by this question yourself and have an answer easily: Is it “dangerous” to keep

npx sst start

running for long periods of time, like overnight? Will the websocket traffic or even the connection count toward the Free Tier usage? Or the other way around: Should I end my debug session without delay? 🙏

Hi @Frank and @Jay, is it possible to build docker image runtime and specify that for the Lambda?

Hi all, when working locally I notice that the permissions assumed by utilities like the AWS-SDK are those of the local system user, and not the ServiceRole that has been assigned to the AWS resource when the stack is constructed. I think the potential here is that local debug works, and the full deploy doesn't, purely due to a permissions issue. I opened an issue for this here: https://github.com/serverless-stack/serverless-stack/issues/363. Is this something of interest?

Hey guys - is there a plan to support Kinesis Data Streams for DynamoDB? Currently the framework only supports the old style streams

This relates to S3 trigger-enabled Lambda

How does SST run Go lambdas locally? Can the mechanism expand to include custom lambda runtimes?

I put together a POC for local AWS Xray support when running

npx sst start

. Would this be of any use for closing serverless-stack#123?

I'm seeing an odd occurence when using a cron-based lambda with SST. When I fire up SST start, the target lambda gets hit multiple times, it almost seems to be that the SST somehow initiates a separate cron invocation in addition to the eventbridge rule/cron. The rate is deliberately set quite low - (every 5 minutes) so invocations don't overlap when debugging. Has anyone seen anything similar?

Hey @Dillon Peterson, do you have

bundle

turned off for you functions?

I have a V1 API backed lambda that I'm running with

sst start

in VSC using debug but it's not hitting any breakpoints. What can I do to track down why?

Silly question here…. how SST compares with Serverless Framework? Does anybody used SLS and moved to SST? Want to hear some feedback/comments/ideas. I been using SLS for 5 years now, and I’m pretty happy with it (not when things fails on the CF side 😂 ).

Is it possible to have more than one entrypoint (index.ts) ?

When SST notices a change in the stack definition it rebuilds and asks for the user to press enter to redeploy. When running SST via a VSC debug session, it doesn't seem to accept the enter press. I guess this is probably to do with the debug window not sending the keyboard event to the process, so I need to restart the entire debug process. Is there any way anyone knows of that I can get around this?

Hi. The YouTube "How to Build a Complete Serverless App on AWS Using SST by Frank Wang" is marked as content made for children and because of that, the "SAVE" to playlist button is disabled.

A nice little helpful feature would be a command line option for sst that outputs the supported version of @aws-cdk/core and @aws-cdk/assert

I have an existing sls application (an app not create by

sst start

),how can i benefit from sst? Is it even possible?

Serverless rules are are now in public preview. (A set of rules to validate IAC template against best practices) https://github.com/aws-samples/serverless-rules

feedback time: just deployed a new project to one of my clients using SST (they use terragrunt/tf for a lot of the stuff which is a bit painful w/ lambda) and their devs amazed are at how good the DX is + how easy to define and understand a stack is 🙌 good stuff, thank you for the awesome job team! In particular to @Frank that’s been so supportive since always. Hopefully will get them using SEED to support the work you guys do there as well!

Just curious (unrelated to the above question), if the stack contains non-serverless part (say I want to have my Beanstalk managed under CDK; and my serverless lambda managed under CDK as well — here SST may come into the picture), does SST still have a leg in this case? • does it have any local debugging mechanism to the non-serverless part? • Does it have any CI/CD value in this hybrid app stack?

In short, we want to isolate deployments in a CI user. But for security, we’d prefer not to have this user be bale to assign higher privilege to itself (which would happen if it had full IAM createRole access). One solution is to add a permissions boundary to policy to this CI user, so that it wouldn’t be able to create users without a permission boundary as well. In the documentation above, the default

Allow

is to all actions on all resource, but it could as well be restricted to whatever CI needs to do in this case

Is .env the recommended way to add “custom context” in sst? https://docs.aws.amazon.com/cdk/latest/guide/context.html

Hey Guys, Is there a way to add a layer while declaring lambdas with

sst.Api

. I see the option in

defaultFunctionProps

but don't see an option to attach it separately to functions.