Error creating ManagedZone due to timeout

[jnahelou]

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
• If an issue is assigned to the "modular-magician" user, it is either in the process of being autogenerated, or is planned to be autogenerated soon. If an issue is assigned to a user, that user is claiming responsibility for the issue. If an issue is assigned to "hashibot", a community member has claimed the issue already.

Hi,

Sometime private DNS zone creation fail with the error :

Error: Error creating ManagedZone: googleapi: Error 409: The resource 'entity.managedZone' named 'XXXXX' already exists, alreadyExists

After debugging, I found the error is due to timeout during the POST request (sendRequestWithTimeout(config, "POST", project, url, obj, d.Timeout(schema.TimeoutCreate)) ) which force Terraform to retry the operation :

2019-11-27T16:48:22.584+0100 [DEBUG] plugin.terraform-provider-google-beta_v2.17.0_x4: 2019/11/27 16:48:22 [DEBUG] Dismissed an error as retryable based on googleapis.com target: Post https://www.googleapis.com/dns/v1beta2/projects/YYYYY/managedZones?alt=json: net/http: request canceled (Client.Timeout exceeded while awaiting headers)
2019-11-27T16:48:22.584+0100 [DEBUG] plugin.terraform-provider-google-beta_v2.17.0_x4: 2019/11/27 16:48:22 [TRACE] Waiting 500ms before next try

From

func isRetryableError(err error, retryPredicates []func(e error) (bool, string)) bool {

        // These operations are always hitting googleapis.com - they should rarely
        // time out, and if they do, that timeout is retryable.
        if urlerr, ok := err.(*url.Error); ok && urlerr.Timeout() {
                log.Printf("[DEBUG] Dismissed an error as retryable based on googleapis.com target: %s", err)
                return true
        }

My actual workaround is to run twice the tf apply...

Is-it possible to add an attribute in the provider to increase the http client timeout in google-beta/config.go waiting for an asynchrone creation model ?

        // Each individual request should return within 30s - timeouts will be retried.
        // This is a timeout for, e.g. a single GET request of an operation - not a
        // timeout for the maximum amount of time a logical request can take.
        client.Timeout, _ = time.ParseDuration("30s")

Terraform Version

Terraform v0.12.10
terraform-provider-google-beta_v2.17.0

Affected Resource(s)

• google_dns_managed_zone

Terraform Configuration Files

provider "google-beta" {
  project = "YYYYYY"
  version = "2.17.0"
}

variable "private_dns_zones" {
  type        = list(string)
  description = "List of all private domains"
  default     = ["sub1.mydomain.com", "sub2.mydomain.com", "sub3.mydomain.com", "sub4.mydomain.com", "sub5.mydomain.com", "sub6.mydomain.com", "sub7.mydomain.com"]
}


resource "google_dns_managed_zone" "priv_sub_domain" {
  provider    = google-beta
  count       = length(var.private_dns_zones)
  name        = replace(var.private_dns_zones[count.index], ".", "-")
  dns_name    = "${var.private_dns_zones[count.index]}."
  description = "Managed by terraform"

  visibility = "private"

  peering_config {
    target_network {
      network_url = "https://www.googleapis.com/compute/v1/projects/XXXX/global/networks/XXXX"
    }
  }
}

Steps to Reproduce

Run this sample multiple time

Link to https://github.com/GoogleCloudPlatform/magic-modules/issues/2756

[nat-henderson]

That is surprising. I did not know that there were GCP APIs that had synchronous, long-running operations with side-effects. :)

I'll see if there's something I can do that won't alter the behavior of other resources.

[nat-henderson]

Thanks for the in-depth bug report, that was very helpful.

[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 feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. If you feel I made an error 🤖 🙉 , please reach out to my human friends 👉 [email protected]. Thanks!