PostgreSQL provider is not using lazy connections

[estahn]

Terraform Version

Terraform v0.12.24
+ provider.aws v2.36.0
+ provider.kubernetes v1.10.0
+ provider.mysql v1.9.0
+ provider.postgresql v1.6.0
+ provider.rabbitmq v1.3.0
+ provider.random v2.2.1

Affected Resource(s)

• provider.postgres

Terraform Configuration Files

provider "postgresql" {
  version  = "~> 1.6"
  alias    = "foo"
  host     = aws_db_instance.foo.address
  username = var.foo_username
  password = var.foo_password
  connect_timeout = 10
  superuser = false
}

Debug Output

aws_db_parameter_group.custom-notmidship: Importing from ID "custom-notmidship"...
aws_db_parameter_group.custom-notmidship: Import prepared!
  Prepared aws_db_parameter_group for import
aws_db_parameter_group.custom-notmidship: Refreshing state... [id=custom-notmidship]
Error: Error initializing PostgreSQL client: error detecting capabilities: error PostgreSQL version: dial tcp 172.31.31.5:5432: connect: operation timed out
  on /myproject/providers.tf line 89, in provider "postgresql":
  89: provider "postgresql" {

Expected Behavior

Running an import action (or any action for that matter) which is not touching posgres resources should not trigger the initialisation of the postgres provider.

Example:

 terraform import aws_db_parameter_group.custom-notmidship custom-notmidship

Actual Behavior

The postgres provider tries to connect to the database.

Steps to Reproduce

Please list the steps required to reproduce the issue, for example:

1. Define postgres provider, see above
2. Run terraform import ...

[mced]

@cyrilgdn Do you have any news on this or a workaround except commenting postgresql providers when importing a resource ?