https://serverless-stack.com/ logo
Join Slack
Powered by
I was looking in the docs but I don’t see steps fo...
# sst
m
I was looking in the docs but I don’t see steps for running sst in an AWS code pipeline. There are some things to address such as sst build outputting to .build/cdk.out but the deploy stage looks for basepath ./cdk.out for example. Is there a good resource for this? Maybe I am messing something up?
f
Hi @Matthew Purdon are you looking to use the CDK pipeline construct?
m
yeah
f
I see. You can do 
sst cdk synth
and that will just run the native 
cdk synth
command. That should spits out 
./cdk.out
(make sure update to the latest SST version ie. v0.10.4)
I haven’t looked into 
CdkPipeline
much. But that’s something we do want to support.
m
yeah, it’s like the other side of the equation, sst makes local dev super great but the pipeline makes the infra part great and effortless to deploy
I think @Michael Clifford mentioned something about flags as well so that we could deploy either the code stack or the pipeline stack (or both)
I see in the docs you recommend two repos but it would be good to support the mono-repo strategy as well
f
Yup, lemme put it down in the discussion.
m
thanks!
Is there anything more you need from us?
f
Yeah, when we get around to look at this, I might have some questions around ur setup and maybe have you guys try out the beta version for it.
Oh btw, just wanted to mention that CDK pipeline deploys all stacks in a CDK app concurrently while respecting inter-stack dependencies. Where as if u were to run 
cdk deploy
, the stacks are deployed one after another, hence much slower.
That’s been one of the attractive features of the pipeline.
SST deploys all stacks concurrently by default.
So there isn’t much to gain deployment speed wise by using the CDK pipeline.
m
yeah, that makes total sense
Open in Slack
PreviousNext
Powered by