This message was deleted.

I think pulumi is waiting for a full delete to happen, not just to schedule the delete and some resource do take a long time to delete. If this is an issue maybe worth raising an issue to discuss it on the aws repo. Might be there's a way to configure the provider to be faster here already, else that seems a reasonable thing to discuss.

Yeah this seems somewhat normal to me. ECS takes a little while to drain/remove the tasks, which is a prereq to fully deleting the service. You can double check for yourself though if you delete the service from the CLI/UI