🚨 OpenTofu Error: 'Provider configuration not present' — Cause, Fix, and Troubleshooting Guide

What this error means

Every resource in state is tied to a provider configuration. If you delete a resource from config but its provider block (often an aliased one) is also gone, OpenTofu still needs that provider to destroy the leftover state object — and cannot find it:

Error: Provider configuration not present

To work with aws_instance.legacy (orphan) its original provider configuration
at provider["registry.opentofu.org/hashicorp/aws"].west is required, but it has
been removed. This occurs when a provider configuration is removed while objects
created by that provider still exist in the state. Re-add the provider
configuration to destroy aws_instance.legacy (orphan), after which you can
remove the provider configuration again.

When it appears

• tofu plan/apply fails with Provider configuration not present.
• The message names an orphaned resource and the (aliased) provider it needs.
• Appears after deleting a provider block or removing a module that defined an aliased provider.
• The resource still exists in state even though its config is gone.

Configuration causes

• Provider block removed while resources it created remain in state.
• Aliased provider deleted — the .west/.replica config was removed but its resources weren’t destroyed.
• Module removed that supplied a configuration_aliases provider.
• Refactor order — config deleted before the resources were destroyed.

Validating the configuration

Read the orphaned resource address and provider alias from the error:

tofu plan 2>&1 | sed -n '1,15p'

Confirm the resource is still in state and which provider it references:

tofu state list | grep legacy
tofu state show 'aws_instance.legacy'

Check whether the provider/alias still exists in config:

grep -rn 'provider "aws"' *.tf

Resolution

Re-add the provider configuration, apply to destroy the orphan, then remove the provider block again:

provider "aws" {
  alias  = "west"
  region = "us-west-2"
}

tofu apply   # destroys aws_instance.legacy using the restored provider

Or remove the object from state if the real infrastructure is already gone (does not touch the cloud):

tofu state rm 'aws_instance.legacy'

Verify the plan is clean afterward:

tofu plan

Keeping definitions valid

• Destroy resources before deleting the provider that manages them; order matters in refactors.
• tofu state rm only forgets the object — use it only when the real resource is already deleted, or you will orphan live infrastructure.
• Aliased providers defined in modules must outlive the resources they created.
• Prefer restoring the provider and running a real destroy over state rm when the resource may still exist.

Related configuration errors

• OpenTofu Error: ‘Invalid provider configuration’ — Cause, Fix, and Troubleshooting Guide
• OpenTofu Error: ‘Invalid legacy provider address’ — Cause, Fix, and Troubleshooting Guide
• OpenTofu Error: ‘Provider does not support resource type’ — Cause, Fix, and Troubleshooting Guide