MySQL Database Instance with Private Network VPC IP only suddenly gets recreated and fails

[Starefossen]

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.

Terraform Version

Terraform v0.11.13
+ provider.google v2.5.1
+ provider.helm v0.9.1
+ provider.kubernetes v1.6.2
+ provider.local v1.2.1
+ provider.random v2.1.1

Affected Resource(s)

• google_sql_database_instance

Terraform Configuration Files

resource "google_compute_network" "net" {
  name             = "my-network"
}

resource "google_sql_database_instance" "db" {
  name             = "my-db"
  database_version = "MYSQL_5_7"
  region           = "eu-north1"

  settings {
    tier = "db-f1-micro"

    ip_configuration {
      private_network = "${google_compute_network.net.self_link}"
    }
  }
}

Apply Output

google_sql_database_instance.db: Modifying... (ID: my-db)
  settings.#:                                    "1" => "1"
  settings.0.activation_policy:                  "ALWAYS" => "<computed>"
  settings.0.availability_type:                  "" => "<computed>"
  settings.0.backup_configuration.#:             "1" => "<computed>"
  settings.0.crash_safe_replication:             "false" => "<computed>"
  settings.0.disk_autoresize:                    "true" => "true"
  settings.0.disk_size:                          "10" => "<computed>"
  settings.0.disk_type:                          "PD_SSD" => "<computed>"
  settings.0.ip_configuration.#:                 "1" => "1"
  settings.0.ip_configuration.0.ipv4_enabled:    "false" => "<computed>"
  settings.0.ip_configuration.0.private_network: "projects/my-project/global/networks/my-net" => "https://www.googleapis.com/compute/v1/projects/my-project/global/networks/my-apps"
  settings.0.location_preference.#:              "1" => "<computed>"
  settings.0.pricing_plan:                       "PER_USE" => "PER_USE"
  settings.0.replication_type:                   "SYNCHRONOUS" => "SYNCHRONOUS"
  settings.0.tier:                               "db-f1-micro" => "db-f1-micro"
  settings.0.version:                            "1" => "<computed>"

Error: Error applying plan:

1 error(s) occurred:

* google_sql_database_instance.db: 1 error(s) occurred:

* google_sql_database_instance.: Error, failed to update instance settings for : googleapi: Error 400: Missing parameter: Settings version., required

Expected Behavior

As per documentation this is how the private network should be configured so the expection is that this should work fine.

Actual Behavior

The instance is created but the second time you run terraform apply terraform says the private_network name had changed and tries to alter it inline and fails.

Steps to Reproduce

1. terraform apply
2. terraform apply <= this will fail

Important Factoids

As a workaround, replacng the https://www.googleapis.com/compute/v1/ part of the VPC network self_link seams to solve the problem.

  settings {
    …
    ip_configuration {
      private_network = "${replace("${google_compute_network.net.self_link}", "https://www.googleapis.com/compute/v1/", "")}"
    }
  }

References

• #0000

[rileykarson]

Hey @Starefossen!

Out of curiousity, have you used lifecycle.ignore_changes on the resource? This is likely an occurrence of hashicorp/terraform#18209

Past that, if you haven't, are you able to provide debug logs or a full repro config? It doesn't look like the config provided matches the values in your apply output.

[rileykarson]

Closing as stale, please file again if you encounter this.

[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!