[AWS::Glue::Workflow] - [BUG] - Workflow update fails unnecessarily when adding multiple triggers

[Rizxcviii]

Name of the resource

AWS::Glue::Workflow

Resource Name

No response

Issue Description

Resource handler returned message: "Workflow my_workflow was modified while adding trigger my_trigger to it (Service: Glue, Status Code: 400, Request ID: xxxxxxxx)" (RequestToken: xxxxxxxxxx, HandlerErrorCode: NotUpdatable)

When this happens, when adding multiple triggers, the entire stack fails to update. However, this can be easily resolved by retrying the creation of the trigger. We could add dependencies, but on deletion, we have to wait for each trigger to delete, where each trigger takes roughly 1-2 minutes to delete. For workflows with 100+ triggers, this could take a whole hour.

Expected Behavior

If the trigger fails with reason "HandlerErrorCode: NotUpdatable", include an exponential backoff algorithm to attempt to delete again, up to a certain number of retries. If it still fails with the same code, then fail the stack

Observed Behavior

The stack fails instantly, with no retries.

Test Cases

Add multiple triggers to the same workflow, and deploy at the same time without dependencies. Then delete the stack, the workflow fails to delete

Other Details

No response