Fix ERROR_DIRECTORY_NOT_RM (0x1A93) on Windows Server

Cause #1: You're pointing to a plain directory, not a Resource Manager

The most common reason you're seeing ERROR_DIRECTORY_NOT_RM (0X00001A93) is that you fired off a command that expects a Resource Manager (RM) path, but you gave it a regular folder. A Resource Manager is a special construct tied to the filesystem that handles transactions and backups. If you run something like fsutil resource or wbadmin against a normal directory, Windows will hit you with this exact error.

For example, I've seen admins run this:

fsutil resource setautoreset true C:\Backup\Folder

That fails because C:\Backup\Folder isn't an RM. The RM lives at the volume root, not in a subdirectory. So the fix is to either use the volume root, or first create an RM in that directory if that's what you're trying to do.

How to fix it

If you're just trying to check or set RM properties, target the drive root instead:

fsutil resource info C:\

If you actually need to create an RM in a specific directory for testing or special scenarios, use:

fsutil resource create C:\MyRM

That command will make that directory a Resource Manager, and then you can point your tools at it.

Cause #2: Using a non-Windows backup tool or a mix of APIs

Another common trigger is when a third-party backup application or a custom script tries to use the Volume Shadow Copy (VSS) or Transactional NTFS (TxF) APIs directly, but passes a path that isn't a registered RM. The API call returns ERROR_DIRECTORY_NOT_RM because the directory doesn't have an RM handle associated with it.

This happens a lot with older backup agents that aren't fully compatible with newer Windows versions. I had a client running a 2012-era backup script on Server 2019, and it kept failing with this error. The script was calling CreateTransaction on a folder that had no RM, and boom.

How to fix it

First, check if the backup tool is supported on your OS version. If it's ancient, update it. If you're writing your own PowerShell or C# code, make sure you're using the right Windows API. For simple file copies, you shouldn't be using transactional APIs at all. Use Copy-Item or Robocopy instead.

If you're stuck with a legacy tool, you might be able to create an RM on the directory that the tool is targeting. But honestly, that's a band-aid. The real fix is to update your tooling.

Cause #3: Corrupted Resource Manager metadata

Sometimes the directory on the volume actually is an RM, but the RM metadata got corrupted. This can happen after a hard crash or a failed disk operation. When the RM's metadata is messed up, Windows can't recognize the directory as a valid RM, so you get the same error.

I've seen this on volumes that were improperly dismounted or had power loss during a transaction. The RM directory still exists, but the internal state is hosed.

How to fix it

You have two options: try to repair the RM, or recreate it.

To check the health of the RM, use:

fsutil resource info C:\Path\ToRM

If that fails or shows errors, you can try a repair:

fsutil resource repair C:\Path\ToRM

That command attempts to fix the RM metadata. If it doesn't work, you'll need to remove the RM and recreate it. But be careful—removing an RM deletes any associated transaction logs, which can cause issues with apps relying on it. Do this only if you're sure no sensitive transactions are pending.

fsutil resource delete C:\Path\ToRM

Then recreate it:

fsutil resource create C:\Path\ToRM

Backup any data inside the RM first because the delete operation might make it inaccessible.

Quick-reference summary table

That should cover the vast majority of cases. If none of those work, you might be dealing with a deeper filesystem issue—run chkdsk /f on the volume and see if that clears it up. But honestly, 90% of the time it's just a wrong path or a tool that's past its prime.