Fix Terraform Error: Moved Object Still Exists at Old Address

Quick Answer

# Remove the OLD resource block — keep only the NEW name + moved block
moved {
  from = aws_instance.web
  to   = aws_instance.app
}

resource "aws_instance" "app" {  # ← Only this block exists
  # ...
}
# Do NOT also have: resource "aws_instance" "web" { ... }

The Error

Error: Moved object still exists

  The configuration still has a resource "aws_instance" "web" declared at
  main.tf:10,1-32, so Terraform cannot move it to "aws_instance" "app".
  
  To complete the move, remove the original resource block.

Or:

Error: Cross-resource move statement

  The move statement at main.tf:1,1-3,2 would conflict with the existing
  resource "aws_instance" "web" at main.tf:5,1-7,2.

What Causes This

1. Old resource block still exists — you added a moved block but didn’t delete the old resource block
2. Both old and new blocks exist — you have aws_instance.web AND aws_instance.app in config
3. Applied moved block not cleaned up — the move succeeded but the moved block is still there
4. Conflicting moves — two moved blocks try to move the same resource

Solution 1: Remove the Old Resource Block

The moved block tells Terraform “the resource formerly at A is now at B.” You must delete A’s resource block:

Before (broken):

# main.tf
moved {
  from = aws_instance.web
  to   = aws_instance.app
}

resource "aws_instance" "web" {  # ❌ Still here — causes error
  ami = "ami-abc123"
}

resource "aws_instance" "app" {
  ami = "ami-abc123"
}

After (correct):

# main.tf
moved {
  from = aws_instance.web
  to   = aws_instance.app
}

resource "aws_instance" "app" {  # ✅ Only the new name
  ami = "ami-abc123"
}

terraform plan   # Shows: aws_instance.web has moved to aws_instance.app
terraform apply  # Updates state, no infrastructure changes

Solution 2: Clean Up Applied Moved Blocks

After terraform apply successfully processes the move, you can remove the moved block:

# After apply — this is safe to remove
# moved {
#   from = aws_instance.web
#   to   = aws_instance.app
# }

resource "aws_instance" "app" {
  ami = "ami-abc123"
}

Tip: Keep moved blocks for one release cycle so all team members and CI/CD environments pick up the move, then remove them.

Solution 3: Use terraform state mv Instead

For quick renames without moved blocks:

# Rename in state directly
terraform state mv aws_instance.web aws_instance.app

Then rename the resource block in your .tf files. No moved block needed.

moved vs state mv

Common Move Patterns

Rename a Resource

moved {
  from = aws_instance.web
  to   = aws_instance.application_server
}

Move Into a Module

moved {
  from = aws_instance.web
  to   = module.compute.aws_instance.this
}

Move Between Modules

moved {
  from = module.old_vpc.aws_vpc.main
  to   = module.networking.aws_vpc.main
}

Rename a Module

moved {
  from = module.web
  to   = module.application
}

Move from count to for_each

moved {
  from = aws_instance.web[0]
  to   = aws_instance.web["primary"]
}

moved {
  from = aws_instance.web[1]
  to   = aws_instance.web["secondary"]
}

Hands-On Courses

• Terraform for Beginners on CopyPasteLearn
• Terraform By Example — practical code examples

Conclusion

The “moved object still exists” error means you have both the old and new resource blocks. Delete the old one — the moved block handles the state migration. Clean up moved blocks after all environments have applied the change. For quick solo renames, terraform state mv is simpler; for team workflows, moved blocks are safer because they’re code-reviewed.