Closed gcasilva closed 1 year ago
Do you think an extra flag is needed to wait? For our usecase this is perfect without any flags. Right now we trigger a clean, a custom script wait for all the stacks tobe deleted and then run clean up resources to delete any leftover s3buckets/dynamodb etc.
But what if someone wants to just trigger the delete taskact test clean project_name
and move on to next steps in the CI/CD without having to wait for the stacks tobe deleted completely in all regions. It might be a deal breaker for such usecases.
I guess something like this will be useful. taskcat test clean project_name --wait-for-stacks-delete
Any thoughts?
@gattasrikanth That's a good point, will work on modifying this implementation so user can send it as a parameter like you suggested: taskcat test clean project_name --wait-for-stacks-delete
@gcasilva Thank you so much for this PR; I'm flipping it to draft while you add the flag; flip it to "ready" when you're... ready :)
@andrew-glenn @gattasrikanth made the change to last commit so that if user does not add argument -w or --wait-for-delete on test clean it will behave as is today (without waiting for cloudformation deletion to complete), but if they add the argument -w then taskcat will wait for cloudformation deletion to complete before proceeding. Tested successfully here. Can you please review it?
Overview
Includes Cloudformation waiter for the delete operation. This is needed when we have a use-case to deploy sequential executions of Taskcat that leverages same set of resources as this could cause conflicts. This is explained in this Issue: https://github.com/aws-ia/taskcat/issues/809
Testing/Steps taken to ensure quality
Tested successfully Taskcat sequential deployment using this forked version
Notes
Optional. Caveats, Alternatives, Other relevant information.
Testing Instructions
How to test this PR Start after checking out this branch (bulleted)