Issue with aws_autoscale_notifications

[stack72]

terraform version
Terraform v0.6.2-dev (8ce3d49895095bbb505d61ead7f79985a04172e3)

I am autogenerating my autoscale groups as follows:

resource "aws_autoscaling_notification" "slack_notifications" {
  group_names = [
    "${split(",", module.autoscalegroups.asg_names)}",
  ]
  notifications  = [
    "autoscaling:EC2_INSTANCE_LAUNCH",
    "autoscaling:EC2_INSTANCE_TERMINATE",
    "autoscaling:EC2_INSTANCE_LAUNCH_ERROR",
    "autoscaling:EC2_INSTANCE_TERMINATE_ERROR",
    "autoscaling:TEST_NOTIFICATION"
  ]
  topic_arn = "<myid>"
}

"${split(",", module.autoscalegroups.asg_names)} generates 21 items

The first time I ran this, 21 autoscale groups managed to be setup for notifications. On subsequenet runs, I get the following:

~ aws_autoscaling_notification.slack_notifications
    group_names.#:          "10" => "21"
    group_names.1065378116: "jenkins-master-autoscale-group" => "jenkins-master-autoscale-group"
    group_names.1143610102: "" => "onboarding-api-autoscale-group"
    group_names.1551555067: "" => "zookeeper-autoscale-group"
    group_names.1775765155: "demo-autoscale-group" => "demo-autoscale-group"
    group_names.1825398809: "" => "media-unit-processor-autoscale-group"
    group_names.2001686116: "dataset-processor-autoscale-group" => "dataset-processor-autoscale-group"
    group_names.2181398257: "" => "riak-autoscale-group"
    group_names.2215067414: "internal-api-autoscale-group" => "internal-api-autoscale-group"
    group_names.2232067133: "" => "rundeck-autoscale-group"
    group_names.2544646555: "dash-autoscale-group" => "dash-autoscale-group"
    group_names.2565852587: "docker-autoscale-group" => "docker-autoscale-group"
    group_names.2847475804: "" => "sensu-autoscale-group"
    group_names.2859170366: "" => "moby-autoscale-group"
    group_names.3180059775: "logstash-autoscale-group" => "logstash-autoscale-group"
    group_names.321274248:  "kibana-autoscale-group" => "kibana-autoscale-group"
    group_names.3491778723: "" => "pipeline-orchestrator-autoscale-group"
    group_names.3561705206: "" => "rich-autoscale-group"
    group_names.3712526780: "kafka-autoscale-group" => "kafka-autoscale-group"
    group_names.4014587927: "elasticsearch-autoscale-group" => "elasticsearch-autoscale-group"
    group_names.532516875:  "" => "teamcity-server-autoscale-group"
    group_names.943818252:  "" => "uchiwa-autoscale-group"

Can the resource only deal with 10 items and thus think it needs to generate 11 more?

Paul

[catsby]

Hey @stack72 – the DescribeNotificationConfigurations API returns a paginated result set, basically 1 entry for each ASG+Notification type. We're only calling it once, so we aren't getting the full list.

I have a working patch here, just need to write a test for it. Should land tonight/tomorrow!

[catsby]

Patched in #3043

[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 have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.