RDS for PostgreSQL instance is tainted right after being created with a parameter that requires a reboot

[gutierrezps]

Terraform Version

Terraform v1.10.0
on windows_amd64
+ provider registry.terraform.io/huaweicloud/huaweicloud v1.72.1

Terraform Configuration Files

terraform {
  required_providers {
    huaweicloud = {
      source  = "huaweicloud/huaweicloud"
      version = "1.72.1"
    }
  }
}

provider "huaweicloud" {
  region     = "sa-brazil-1"
  access_key = var.hwc_access_key
  secret_key = var.hwc_secret_key
}

resource "huaweicloud_vpc" "test" {
  name = "vpc-test"
  cidr = "10.0.0.0/16"
}

resource "huaweicloud_vpc_subnet" "test" {
  name       = "subnet-test"
  cidr       = "10.0.0.0/24"
  gateway_ip = "10.0.0.1"
  vpc_id     = huaweicloud_vpc.test.id
}

resource "huaweicloud_networking_secgroup" "test" {
  name                 = "sg-test"
  delete_default_rules = true
}

resource "huaweicloud_rds_instance" "instance" {
  name                = "rds-test"
  flavor              = "rds.pg.n1.large.2.ha"
  ha_replication_mode = "async"
  vpc_id              = huaweicloud_vpc.test.id
  subnet_id           = huaweicloud_vpc_subnet.test.id
  security_group_id   = huaweicloud_networking_secgroup.test.id
  availability_zone   = ["sa-brazil-1a", "sa-brazil-1b"]

  db {
    type     = "PostgreSQL"
    version  = "16"
    password = var.password
  }

  volume {
    type = "CLOUDSSD"
    size = 40
  }

  parameters {
    name  = "cron.timezone"
    value = "America/Sao_Paulo"
  }
}

Debug Output

https://gist.github.com/gutierrezps/fa02621e7fa549b3ab8b7d64f2507bbb

> terraform apply
...
huaweicloud_rds_instance.instance: Still creating... [6m30s elapsed]
huaweicloud_rds_instance.instance: Still creating... [6m30s elapsed]

 Error: error rebooting for RDS instance (**rds_id**): Expected HTTP response code [200 202] when accessing [POST https://rds.sa-brazil-1.myhuaweicloud.com/v3/**project_id**/instances/**rds_id**/action], but got 409 instead.
│ request_id: **request_id**, error message: {"error_msg":"Another operation is being performed on the DB instance or the DB instance is faulty.","error_code":"DBS.200452"}   
│
│   with huaweicloud_rds_instance.instance,
│   on main.tf line 33, in resource "huaweicloud_rds_instance" "instance":
│   33: resource "huaweicloud_rds_instance" "instance" {

> terraform plan
...
Terraform will perform the following actions:

  # huaweicloud_rds_instance.instance is tainted, so must be replaced
-/+ resource "huaweicloud_rds_instance" "instance" {
 ...

Expected Behavior

RDS should be created successfully and saved to the state file.

Actual Behavior

RDS instance is created in the cloud account, the parameter is updated (confirmed in the Console), but since the parameter cron.timezone is only effective upon reboot, Terraform attempted to reboot the instance. The reboot failed (as seen in the error message above), so Terraform considers the RDS instance is tainted and replaces it in the next time terraform apply is executed.

Steps to Reproduce

1. terraform init
2. terraform apply -- it fails to reboot the instance
3. terraform apply -- it shows the RDS resource is tainted and needs to be replaced

Additional Context

cron.timezone parameter from RDS for PostgreSQL instance is only effective upon reboot.