Skip to content

[BUG] - The LoadBalancer resource has ipv4 and ipv6 set as null in the state after a successful creation #635

Description

@aittam

Describe the bug
After creating a Load Balancer resource, the terraform state doesn't contain the ipv4 and ipv6 attributes (see screenshot).

To Reproduce
Steps to reproduce the behavior:

  1. Create a LB resource
  2. Try to access the ipv4 attribute
resource "vultr_load_balancer" "lb" {

  region              = var.region_id
  label               = "${var.cluster_name}-lb"
  balancing_algorithm = "roundrobin"
  vpc                 = var.vpc_id
  attached_instances  = module.cp_instance.*.id

  health_check {
    path     = "/livez"
    port     = 6443
    protocol = "https"
  }
  forwarding_rules {
    frontend_protocol = "tcp"
    frontend_port     = 6443
    backend_protocol  = "tcp"
    backend_port      = 6443
  }
}

resource "azurerm_dns_a_record" "apps" {
  name                = "*.apps.${var.cluster_name}"
  zone_name           = var.domain
  resource_group_name = "my-rg"
  ttl                 = 300
  records             = [vultr_load_balancer.lb.ipv4]
}

Expected behavior
Both ipv4 and ipv6 attributes to be populated correctly.

Screenshots

Image

Desktop (please complete the following information where applicable:

  • OS: MacOS 15.7.2
  • Terraform v1.14.0
  • Provider version v2.27.1

Additional context

Rerunning terraform a second time fetch the informations correctly.

As a workaround we can use a data that depends on the LB resource and collect the informations with a second API call, but that can still potentially fail if it's a timing issue.

Metadata

Metadata

Assignees

Labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions