Error registering targets with target group: InvalidTarget: The following targets are not in a running state and cannot be registered

[ghost]

This issue was originally opened by @edeas123 as hashicorp/terraform#20345. It was migrated here as a result of the provider split. The original body of the issue is below.

---

Terraform Version

Terraform v0.11.11
+ provider.aws v1.54.0

Terraform Configuration Files

resource "aws_alb" "rancher-ctl-host-alb" {
	name = "rancher-ctl-host-alb"
	internal = false
	load_balancer_type = "application"
	ip_address_type  = "ipv4"

	subnets = ["${data.aws_subnet_ids.vpc_subnets.ids}"]
	security_groups = [
		"${data.terraform_remote_state.core.rancher-ctl-host-alb-sg-id}"
	]
}

# create an application load balancer listener 
resource "aws_alb_listener" "rancher-ctl-host-alb-listener" {
	load_balancer_arn = "${aws_alb.rancher-ctl-host-alb.arn}"
	port = 80
	protocol = "HTTP"

	default_action {
    	type             = "forward"
    	target_group_arn = "${aws_alb_target_group.rancher-ctl-host-target-group.arn}"
  	}
}

# create the application load balancer target group
resource "aws_alb_target_group" "rancher-ctl-host-target-group" {
	name = "rancher-ctl-host-target-group"
	port = 8080
	protocol = "HTTP"
	vpc_id = "${data.terraform_remote_state.core.default-vpc-id}"
}

# attach the three hosts to the target group
resource "aws_alb_target_group_attachment" "rancher-ctl-host-target-group-instances" {
  target_group_arn = "${aws_alb_target_group.rancher-ctl-host-target-group.arn}"
  target_id        = "${aws_spot_instance_request.rancher-ctl-host.*.spot_instance_id[count.index]}"
  port             = 8080
  count            = 3
}

Expected Behavior

It should not have given errors applying the plan

Actual Behavior

Error: Error applying plan:

3 error(s) occurred:

* aws_alb_target_group_attachment.rancher-ctl-host-target-group-instances[2]: 1 error(s) occurred:

* aws_alb_target_group_attachment.rancher-ctl-host-target-group-instances.2: Error registering targets with target group: InvalidTarget: The following targets are not in a running state and cannot be registered: 'i-0791f6bee8a082a10'
        status code: 400, request id: 990ee63b-3053-11e9-92bd-4d5e8013e613
* aws_alb_target_group_attachment.rancher-ctl-host-target-group-instances[0]: 1 error(s) occurred:

* aws_alb_target_group_attachment.rancher-ctl-host-target-group-instances.0: Error registering targets with target group: InvalidTarget: The following targets are not in a running state and cannot be registered: 'i-0b8c1f6d35f57c5cb'
        status code: 400, request id: 990f0dbc-3053-11e9-bbd2-3b6b83537945
* aws_alb_target_group_attachment.rancher-ctl-host-target-group-instances[1]: 1 error(s) occurred:

* aws_alb_target_group_attachment.rancher-ctl-host-target-group-instances.1: Error registering targets with target group: InvalidTarget: The following targets are not in a running state and cannot be registered: 'i-033d01746af85be03'
        status code: 400, request id: 990f3424-3053-11e9-aa19-bf12dffa0d2b

Additional Context

The problem seem to be that the alb_target_group_attachment is created before the spot instances are running (although they have been created), and it does not support interpolation which is needed in my case to use the depends_on.

References

• [AWS] Creating of ALB, Target Group and Service in one script causes errors terraform#12634

[nywilken]

Hi @edeas123 sorry to hear you are running into issues applying this plan. Have you tried reaching out to the community forums for assistance?

We use GitHub issues for tracking bugs and enhancements rather than for questions. While we may be able to help with certain simple problems here it’s generally better to use one of the community forums where there are far more people ready to help, whereas the GitHub issues here are generally monitored only by the small set of code maintainers of each repository.

[edeas123]

@nywilken are you saying this does not qualify as a bug?

[bflad]

@edeas123 just to confirm your configuration, do you have wait_for_fulfillment enabled for the aws_spot_instance_request resources? Does a second terraform apply work? If so, this may be a case where we need to tell the aws_lb_target_group_attachment to retry on that specific message for a few minutes to allow the spot instances to go from pending to running.

[edeas123]

@bflad yes to both questions

[nywilken]

Hi @edeas123 my apologies for prematurely closing the issue. In reading the provided information I thought the issue was more of a question around timing issues with the use of spot instances since the expected behavior was "It should not have given errors applying the plan". My apologies for making that assumption. I'll update the labeling on this and reopen.

[jbarrick-mesosphere]

Opened a PR: #8483

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. Thanks!