Why this is a problem?
If the cloudformation stacks are not deleted at the end of a given build, then they become orphaned. Then, it'll continue to accrue cost, and will also count towards AWS resource limits. For ex. we have limits on how many Cloudformation stacks, VPCs, EC2 instances, RDS instances we can create.
So, the orphaned stacks pose a big problem over-time because, once the limits are reached, all the TG builds will start failing.
We need to find out what are the reasons they don't get deleted. This seems to be an intermittent issue.
Following is a sample set of stacks that were orphaned in us-east-2 region.
