customTimeouts not honored for `ecs.Service`

When passing this customTimeouts block to new awsx.ecs.FargateService():

  80   │   {
  81   │     customTimeouts: {
  82   │       create: '30m',
  83   │       update: '30m',
  84   │       delete: '30m',
  85   │     },
  86   │   }

Pulumi still reports that the operation times out after 10 minutes:

Diagnostics:
  aws:ecs:Service (development-root-fargate-service):
    error: Plan apply failed: 1 error occurred:
        * creating urn:pulumi:development::infra::awsx:x:ecs:FargateService$aws:ecs/service:Service::development-root-fargate-service: timeout while waiting for state to become 'true' (last state: 'false', timeout: 10m0s)

Judging by this comment (https://github.com/pulumi/pulumi/issues/2655#issuecomment-495302984), it seems like what might be required is better surfacing of the fact that custom timeouts are not supported for this resource type.

[lukemundy-vgw]

Just wanted to bump this issue as it’s causing a small amount of pain for us during deployments. Our ECS services are typically pretty quick to deploy (less than 2 minutes) and so I’d like to set a pretty short timeout on them (something like 5 minutes) so that if there are issues with the service stabilising, we get much quicker feedback than having to wait for the default timeout of 15 minutes, or an engineer noticing that deployment is taking longer than usual.

[lukehoban]

Ok, so I found the issue here - essentially, the underlying ecs_service doesn’t actually support CustomTimeouts.

CustomTimeouts support was added in the upstream provider in https://github.com/hashicorp/terraform-provider-aws/pull/10452 (for delete) and in https://github.com/hashicorp/terraform-provider-aws/pull/25641 (for create and update).