[ISSUE] Issue with renaming databricks_catalog resource. #3818
Description
Configuration
before running the code, we had someone manually create a catalog from a delta share called "prod2_share". the goal here is 1) import that catalog into terraform then (2) rename it then (3) create a workspace binding for it
resource "databricks_catalog" "prod2_share" {
name = "prod2"
share_name = "prod-share"
isolation_mode = "ISOLATED"
provider_name = "aws:us-east-1:xxxxx-xxxx-xxx-xxx-xxxxxxxxx"
}
import {
provider = databricks.workspace
to = databricks_catalog.prod2_share
id = "prod2_share"
}
resource "databricks_catalog_workspace_binding" "prod2_shared" {
provider = databricks.workspace
securable_name = databricks_catalog.prod2_share.name
workspace_id = "XXXXXX"
}plan from the above showed....
# databricks_catalog.prod will be updated in-place
# (imported from "prod2_share")
! resource "databricks_catalog" "prod2_share" {
+ force_destroy = false
id = "prod2_share"
isolation_mode = "ISOLATED"
metastore_id = "xxxxxxx-xxx-xxx-xxxx-xxxxxx"
! name = "prod2_share" -> "prod2"
owner = "xxxxxxx-xxxx-xxxx-xxxx-xxxx"
provider_name = "aws:us-east-1:xxxxx-xxxx-xxx-xxx-xxxxxxxxx"
share_name ="prod-share"
}
# databricks_catalog_workspace_binding.prod2_shared will be created
+ resource "databricks_catalog_workspace_binding" "prod2_shared" {
+ binding_type = "BINDING_TYPE_READ_WRITE"
+ id = (known after apply)
+ securable_name = "prod2"
+ securable_type = "catalog"
+ workspace_id = xxxxxxxxx
}
Expected Behavior
- on terraform apply, the above plan works as intended, it renames the catalog and all is good
Actual Behavior
- can see in apply it thinks it did the modification
databricks_catalog.prod2_share[0]: Modifying... [id=prod2_share]
databricks_catalog.prod2_share[0]: Modifications complete after 0s [id=prod2_share]
- then throws this error...
│ Error: Provider produced inconsistent final plan
│
╷
│ When expanding the plan for
│ databricks_catalog_workspace_binding.prod2_shared[0] to include new values
│ learned so far during apply, provider
│ "registry.terraform.io/databricks/databricks" produced an invalid new value
│ for .securable_name: was cty.StringVal("prod2"), but now
│ cty.StringVal("prod2_share").
│
│ This is a bug in the provider, which should be reported in the provider's
│ own issue tracker.
- aftewards i checked if the imported catalog was renamed, and it was NOT. This i assume is the root issue here, which is renaming it did nothing.
- obviously i should of imported first, then done follow up PRs to rename/workspace bindings, but not sure if that would fix, considering it didnt rename the catalog despite it thinking it did. This also resulted in a bad tf state, where it thinks the catalog has a name now that it does not.
Steps to Reproduce
- in one apply, import a delta share catalog, rename it, and create a workspace binding from it
Terraform and provider versions
1.49