provider/aws: route53_record - "fqdn" is missing during first apply

[jamestoyer]

I seem to be having a problem displaying the fqdn property on a Route 53 record when using it with an output. The output variable does not appear at the end of the apply stage with all my other outputs.

I have the output set up as

output "web_fqdn" {
  value = "${aws_route53_record.web.fqdn}"
}

And I have the following Route 53 record set up as

resource "aws_route53_record" "web" {
   zone_id = "${var.dns_domain_zone_id}"
   name = "web.${var.dns_domain}"
   type = "CNAME"
   ttl = "300"
   records = ["${aws_elb.web.dns_name}"]
}

The apply stage was run using Atlas (and the rather awesome GitHub integration), so it's version 0.5.3, as far as I can see.

The actual record was created correctly, and I've verified it exists. So far I've only tested this on the creation of a record, and not subsequent applies, i.e. when the record already exists and it's not been updated.

[jamestoyer]

Just to post an update.

I've just done an apply on Atlas, which used v0.6.0-dev. This time I was not updating the Route 53 records (there were other changes to an ELB), and the output appeared as expected.

[radeksimko]

@jamestoyer Thanks for reporting this.

I can confirm that fqdn is missing in the output in the first apply run.
Here's a snippet of code I was testing:

variable "dns_domain" {
  default = "notexample.com"
}

resource "aws_route53_zone" "primary" {
  name = "${var.dns_domain}"
}

resource "aws_route53_record" "web" {
  zone_id = "${aws_route53_zone.primary.id}"
  name = "web.${var.dns_domain}"
  type = "CNAME"
  ttl = "300"
  records = ["${aws_elb.web.dns_name}"]
}

resource "aws_elb" "web" {
  name = "foobar-terraform-elb"
  availability_zones = ["us-west-2b", "us-west-2c"]

  listener {
    instance_port = 80
    instance_protocol = "http"
    lb_port = 80
    lb_protocol = "http"
  }
}

output "web_fqdn" {
  value = "${aws_route53_record.web.fqdn}"
}

output "web_name" {
  value = "${aws_route53_record.web.name}"
}

after 1st terraform apply:

Apply complete! Resources: 3 added, 0 changed, 0 destroyed.

...

Outputs:

  web_name = web.notexample.com

2nd terraform apply:

Apply complete! Resources: 0 added, 0 changed, 0 destroyed.

Outputs:

  web_fqdn = web.notexample.com
  web_name = web.notexample.com

[sheldonh]

Is this about aws_route53_record not waiting long enough to confirm that the resource has been created? This is in my way, and I'm keen to take a poke at it.

[catsby]

I opened #2528 to patch this; TL;DR we typically call update after create, and update typically calls read.

Here we just quit after create, so calling read at the end populates the fqdn attribute

[catsby]

#2528 was merged, thanks for reporting this issue!

[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.