Deleting (part of) CloudBuilder deployment

Question

Hi,

Scenario: Multiple ArcGIS Enterrpise (10.8, 10.8.1) deployments using CloudBuilder for AWS Windows app;

Goal: Fully delete one of the instances (ec2, associated host storage and S3, related components and not delete the load balancer, vpc, and anything else not related to that first deployment.

We started with one (10.8), then later added another (10.8.1) and the third (10.8.1) was thankfully our last! 
We've moved everything off that first instance, traffic is cut so it's ready to go.

The trio all live in a single VPC and share a elastic loadbalancer, but have separated security groups and separate S3 storage; the newer two were blessed with external IP addresses while the first one wasn't. We tried really hard to deploy them as independent as we possibly could.

So i fired up the CloudBuilder for AWS (Windows) app and see my trio; I'm tempted to just click that trash can and hold my breath, but the alarm bells sound off in my head.
I went hunting for documentation but find none.

How should i go about deleting that first instance and all of its related parts, but not (accidentally?) delete the load balancer, the vpc and other stuff that isnt a part of the first instance? Will clicking that trash can do exactly this?

Thanks,

Derek

p.s.: yes, the other two deployments are hosting production services 🙂

ChristopherPawlyszyn · Accepted Answer

If you login to the AWS Console and go to CloudFormation, you should be able to see a distinct stack created for each site. Clicking a stack and going to the Resources tab should allow you to see what resources were created as part of that stack, and which would be destroyed if you delete the entire stack. Hopefully that helps you identify a path forward since the Cloud Builder application pulls information from the CloudFormation APIs. If the VPC, ELB, and other resources were provisioned outside of the stack you are looking to delete, it should be safe to delete that stack.

One thing to keep in mind is that the associated DynamoDB table(s) and S3 buckets won't be removed when you delete the stack and will require manual cleanup for those resources. They should be named appropriately to include each stack name so you can identify those resources.