core: Add resource IDs to apply-errors + prevent error duplication

[radeksimko]

This is a proposal for fixing #2801 and similar with a more systematic approach.

Current state:

aws_route53_record.www: Creating...
  fqdn:               "" => "<computed>"
  name:               "" => "www.notexample.com"
  records.#:          "" => "1"
  records.3619153832: "" => "127.0.0.1"
  ttl:                "" => "300"
  type:               "" => "A"
  zone_id:            "" => "WRONGID"
aws_route53_record.www: Error: 1 error(s) occurred:

* NoSuchHostedZone: No hosted zone found with ID: WRONGID
    status code: 404, request id: [d7febede-3063-11e5-9472-2971854b3f91]
Error applying plan:

1 error(s) occurred:

* NoSuchHostedZone: No hosted zone found with ID: WRONGID
    status code: 404, request id: [d7febede-3063-11e5-9472-2971854b3f91]

Terraform does not automatically rollback in the face of errors.
Instead, your Terraform state file has been partially updated with
any resources that successfully completed. Please address the error
above and apply again to incrementally change your infrastructure.

this PR is changing it into the following (no duplicate errors and resource ID mentioned):

aws_route53_record.www: Creating...
  fqdn:               "" => "<computed>"
  name:               "" => "www.notexample.com"
  records.#:          "" => "1"
  records.3619153832: "" => "127.0.0.1"
  ttl:                "" => "300"
  type:               "" => "A"
  zone_id:            "" => "WRONGID"
Error applying plan:

1 error(s) occurred:

* aws_route53_record.www: NoSuchHostedZone: No hosted zone found with ID: WRONGID
    status code: 404, request id: [1db1c357-3064-11e5-8c8a-916e2ed8af08]

Terraform does not automatically rollback in the face of errors.
Instead, your Terraform state file has been partially updated with
any resources that successfully completed. Please address the error
above and apply again to incrementally change your infrastructure

Here's a simple way to trigger an error (and see the difference):

provider "aws" {
    region = "us-west-2"
}

resource "aws_route53_record" "www" {
   zone_id = "WRONGID"
   name = "www.notexample.com"
   type = "A"
   ttl = "300"
   records = ["127.0.0.1"]
}

git blame command/hook_ui.go only brought up @mitchellh , so I think he would be the best candidate for reviewing.

The other option would be to keep the error printing in UiHook which has the advantage of more "real-time" feedback, but in that case we should:

1. Print it in red (it is now bold white)
2. Somehow filter these errors out in eval_apply.go so it's not duplicated.

FYI @failshell @catsby

[radeksimko]

In fact it's not fixing #2801 (yet) since it's just addressing apply, not refresh, I'll think about refresh too. Maybe similar approach can be taken.

[radeksimko]

Ok, same approach for refresh applied, now I need your votes/opinion about reporting via UiHook or aggregating errors and printing at the end.

I'm slightly more inclined to hooks, but both solutions have different pros/cons.

[catsby]

I only tested this with the Route 53 example from 2801, but overall 👍 to this idea. The change seems small enough, but word from @mitchellh or @phinze may be prudent

[phinze]

This is nice! LGTM

[radeksimko]

So you both prefer error aggregation over real-time errors then?
Just checking you did not overlook my question... :)

[phinze]

Ah thanks for pointing that out - I had overlooked the impl question.

I think getting resource IDs into the error message is an important enough win that it's better to land this as-is and we can always circle back if we find that waiting for the errors to output becomes annoying. 👍

[radeksimko]

@phinze Agreed, it can be revisited later.

[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.