monitor setting new_host_delay never takes effect

[christopherriley]

terraform 0.11.8
datadog provider 1.3.0

datadog monitor will never be created/updated with non-default value for new_host_delay - new_host_delay is ALWAYS set to the default value of 300

Terraform Version

xxx:/conf/site# terraform -v
Terraform v0.11.8
+ provider.aws v1.35.0
+ provider.datadog v1.3.0

Affected Resource(s)

• datadog_monitor

Terraform Configuration Files

resource "datadog_monitor" "mailgun_endpoint_health_check" {
  name = "Mailgun Endpoint Health Check (${var.site})"
  type = "service check"
  tags = "${concat("${local.monitor_tags}", "${list("service:mailgun")}")}"

  query = <<QUERY
  "supernova.health_check.mailgun.endpoint".over("site:${var.site}").by("site","endpoint","orgid").last(4).count_by_status()
QUERY

  message = "the_message"
  escalation_message = "the_escalation_message"

  thresholds {
    warning  = "${var.mailgun_warn_threshold}"
    critical = "${var.mailgun_critical_threshold}"
    ok       = "${var.mailgun_ok_threshold}"
  }

  include_tags      = true
  locked            = false
  notify_audit      = false
  notify_no_data    = true
  no_data_timeframe = 45
  renotify_interval = 30
  timeout_h         = 0
  new_host_delay = 0

Expected Behavior

datadog monitor should be created/updated with new_host_delay = 0

Actual Behavior

datadog monitor is always created/updated with new_host_delay = 300 (300 being the default)

Steps to Reproduce

Please list the steps required to reproduce the issue, for example:

1. terraform apply
2. terraform apply
3. terraform apply
   ... etc

References

• probably somehow related to https://github.com/terraform-providers/terraform-provider-datadog/issues/87

[masci]

Fixed in #101