Hi all, I am nearing conversion of my application to SST, but have the following question: I have a deployment pipeline which invokes

sst deploy

at the end (once the code is tested etc). Now the issue I run into is the following:

Integration-api-integration-MyTable already exists

so

Failed to deploy the app

Is it not possible to continually deploy the application? What would be best practice here? Apologies if this has been covered somewhere but I couldn't find it in the docs under the deployment section (or perhaps I misunderstood).

Certain resources have a removal policy of retain by default. This is for things like databases so they're not automatically deleted on stack destruction

Thanks @thdxr, just so I am clear: Suppose I have had the app running for some time in development. I have databases that are populated already. I want to make updates to the deployment (perhaps I add another lambda function), but retain this table with live data. Would setDefaultRemovalPolicy still apply? Because I am not looking to destroy resources, simply add another lambda function (or indeed remove some code). Apologies if the question is basic I am pretty new to AWS/Serverless Stack.

What version of sst are you on?

I am on the latest version. Things seem to be working now, but the gist I get is: If I try and deploy a stack, if for any reason it fails, then redeploying a new stack runs into the "already exists" issue, rather than over-writing.

@Fazi yeah, when creating a new stack fails half way, the deployment is automatically rolled back, and the created resources gets removed.