aws/route53: Input for record.*.name != record.*.name on the output

[radeksimko]

This is probably low-prio + easy to do, but it's more important to discuss it here.

Look at the following code and tell me what would you expect the output to be?

resource "aws_route53_zone" "primary" {
   name = "example.radeksimko.com"
}

resource "aws_route53_record" "sample" {
  zone_id = "${aws_route53_zone.primary.id}"
  name = "sample"
  type = "CNAME"
  ttl = "60"
  records = ["www.terraform.io"]
}

output "name" {
  value = "${aws_route53_record.sample.name}"
}

I would definitely expect ${aws_route53_record.sample.name} to be sample, not the full domain (sample.example.radeksimko.com).
User will most likely want to use the full domain elsewhere instead of concatenating it (so it should really stay in the output), but I'm just not sure about the name here.

How about something like aws_route53_record.sample.full_name or aws_route53_record.sample.domain or aws_route53_record.sample.fqdn?

[catsby]

This is partially the result of #969 , where we allow people to not have to use the fully qualified name. Here, name represents "DNS domain name"

• http://docs.aws.amazon.com/Route53/latest/APIReference/API_ChangeResourceRecordSets_Requests.html#API_ChangeResourceRecordSets_RequestBasicSyntax

Prior to #969 , you would have needed to specify the fully qualified name, so you would then also get sample.example.radeksimko.com. At least then, it wouldn't be surprise...

In theory, we could add a new computed schema element that is ignored in configuration, and is set from the name value here? Does that sound sufficient?

[radeksimko]

This is partially the result of #969 , where we allow people to not have to use the fully qualified name.

yeah, except it's effectively now useless as it triggers the update all the time... 😉

I do have to use the full FQDN everywhere anyway if I want consistent behaviour and not recreate the record each time I run terraform apply :)

In theory, we could add a new computed schema element that is ignored in configuration, and is set from the name value here? Does that sound sufficient?

That's basically what I meant and what I was suggesting. 👍

The whole confusion just comes from the fact that input.name != output.name - the same attribute on input may not always equal the attribute on output.

[radeksimko]

This has been fixed in #1279

[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.