Terraform-provider-aws: `aws_cloudfront_distribution` resource cannot properly parse variable of type list

[saidmasoud]

Community Note

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
• Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
• If you are interested in working on this issue or have submitted a pull request, please leave a comment

Terraform Version

$ terraform -v
Terraform v0.11.8
+ provider.aws v1.35.0
+ provider.template v1.0.0

Affected Resource(s)

• aws_cloudfront_distribution

Terraform Configuration Files

# Route 53 Module
resource "aws_route53_record" "record1" {
  zone_id = "${aws_route53_zone.base.zone_id}"
  name    = "<redacted>"
  type    = "CNAME"
  ttl     = "300"
  records = ["<redacted>"]
}

resource "aws_route53_record" "record2" {
  zone_id = "${aws_route53_zone.base.zone_id}"
  name    = "<redacted>"
  type    = "CNAME"
  ttl     = "300"
  records = ["<redacted>"]
}

output "cnames" {
    value = ["some.thing.com,${aws_route53_record.record1.name},${aws_route53_record.record2.name}"]
}

# CloudFront distribution
variable "cnames" {
    type = "list"
}

resource "aws_cloudfront_distribution" "reappr-dev" {
.........
aliases = ["${var.cnames}"]
}

Debug Output

The debug output is massive and contains sensitive info. Can be privately shared upon request.

Panic Output

N/A

Expected Behavior

Terraform should have extracted the list from the variable and set the CNAMEs appropriately

Actual Behavior

The AWS API returned a 404:

Error: Error applying plan:

1 error(s) occurred:

* module.cf.aws_cloudfront_distribution.reappr-dev: 1 error(s) occurred:

* aws_cloudfront_distribution.reappr-dev: error updating CloudFront Distribution (ES9QE524IXBN): InvalidArgument: The parameter CNAME contains one or more parameters that are not valid.
    status code: 400, request id: f553e53c-d70e-11e8-a6d2-ff00a0e2fdc0

Terraform does not automatically rollback in the face of errors.
Instead, your Terraform state file has been partially updated with
2018-10-23T17:59:48.581-0400 [DEBUG] plugin.terraform-provider-aws_v1.35.0_x4: 2018/10/23 17:59:48 [ERR] plugin: plugin server: accept unix /var/folders/6d/y54y04v51t76038v3k2nflm40000gn/T/plugin661750983: use of closed network connection
any resources that successfully completed. Please address the error
above and apply again to incrementally change your infrastructure.

Steps to Reproduce

Create the above resources, then apply.

Important Factoids

Manually hard-coding CNAME record into distribution resource works.

One of the CNAMEs in the Route 53 module output is hard-coded due to a cycle error.

[esanchezm]

EDIT: Please ignore, turns out one of the domains I was providing as alias included an underscore character (xx_xx.my-domain.com) and it's not valid to use as CNAME. "Domain names must contain one or more dots (.) and can only include lower case alphanumeric characters, dashes (-), and, optionally, a leading "." to indicate all subdomains of the specified domain, for example, ".example.com".

--

I'm having the same issue with TF v0.12.13 and AWS provider v2.35.0. I wrote a module that creates the CF Distribution like this:

variable "aliases" {
  type        = list(string)
  description = "List of aliases domains for the Cloudfront distribution"
}

resource "aws_cloudfront_distribution" "www-site-cloudfront" {
  ...
  aliases = var.aliases
  ...
}

And I'm getting this error:

Error: error updating CloudFront Distribution (EXXXXXXXXX): InvalidArgument: The parameter CNAME contains one or more parameters that are not valid.
    status code: 400, request id: 1a7bdb33-xxxxxxxxx

[normankong]

I got exactly the same error and luckily after reading @esanchezm post, I should change the CName into lower capital letter. Then the deployment complete successfully. Thanks a lot !