Fix 0X0000170C Duplicate Cluster Name Error

The 30-Second Fix: Check AD for the Stale Object

This error pops up when you try to create a cluster with a name that's already claimed in Active Directory. Maybe an old cluster was decommissioned but the computer object wasn't cleaned up. Or someone re-created a cluster with the same name on a different server.

Open Active Directory Users and Computers on your domain controller (or use ADAC). Expand your domain, then look in the Computers container. Search for your-cluster-name$. If you see a computer object with that name, right-click it and select Delete. After you do, wait about 30 seconds for AD replication (or force it with repadmin /syncall if you're impatient). Then try your cluster creation again.

If you don't see it in Computers, also check the Microsoft Exchange Security Groups or System > Microsoft Exchange > Microsoft System Attendant containers if you have Exchange — clusters sometimes leave stray objects there. But nine times out of ten, it's in Computers.

Real-world trigger: I've seen this most often when someone builds a cluster as a proof-of-concept, deletes it without cleaning up AD, then tries to reuse the same name six months later. The old computer object is still sitting there.

The 5-Minute Fix: Check DNS and Flush the Record

If AD looks clean but the error still shows up, a stale DNS record is your next suspect. Open DNS Manager on your domain controller. Expand your forward lookup zone (usually named after your domain). Look for an A or AAAA record matching your cluster name. If it exists, right-click and Delete it.

Then open PowerShell as Administrator on the server where you're building the cluster. Run:

ipconfig /flushdns

Then clear the cluster's own cached records:

Clear-DnsClientCache

Now try creating the cluster again. If it still fails, it's time to dig deeper — the cluster creation wizard may be caching old data itself.

A quick side note: if you're using Server Core, you won't have the GUI tools. Use PowerShell instead. Run Get-DnsServerResourceRecord -ZoneName "yourdomain.com" -Name "your-cluster-name" to find the record, then Remove-DnsServerResourceRecord to delete it.

The 15+ Minute Fix: Clean Up Cluster Registration in AD and the Cluster Database

Sometimes the cluster name is registered in a hidden place — the cluster's own AD object. Even if you deleted the computer object, an orphaned cluster object can linger. Here's the brute-force approach that always works for me.

First, check if there's a cluster object in AD under CN=Computers,DC=yourdomain,DC=com with objectClass msCluster-Computer. Use ADSI Edit (install it from Remote Server Administration Tools if needed). Connect to your domain's Configuration partition. Navigate to CN=Services, CN=Windows NT, CN=Cluster. Look for an entry matching your cluster name. Delete it.

Now, on the server where you're creating the cluster, open PowerShell as Administrator and run:

Get-Cluster | Remove-Cluster -Force

This removes any leftover cluster configuration from the server's local registry and AD. But note: if you don't have a cluster yet, this command might error out — that's fine, skip to the next step.

Next, clear the DNS record again (just in case the cluster re-created it). Then re-create the cluster with a slightly different name if the original name is still stubborn. I know, it's not ideal, but it works. You can rename the cluster later using Rename-Cluster after the initial creation.

If you're still stuck, check the event log on the server under Applications and Services Logs > Microsoft > Windows > Failover Clustering > Operational. Look for Event ID 1002 or 1074 — they'll tell you exactly which resource is conflicting. Usually it's the Network Name resource.

One last trick: Open adsiedit.msc. Connect to the Domain partition. Expand DC=yourdomain,DC=com, then CN=System, then CN=MicrosoftDNS. Find your zone, locate the stale cluster record, and delete it there. DNS scavenging might not have caught it if it's static. Yes, this is painful, but it's the nuclear option and it always kills the ghost.

After all this, reboot the server. I know it sounds old-school, but I've seen the cluster service hold onto cached credentials that only a reboot clears. Come back up, try the cluster creation again. It'll work.