When This Error Hits
You're setting up a Windows Failover Cluster and you see this error in the Validation Wizard or when creating an IPv6 Tunnel resource. The exact text: ERROR_CLUSTER_INVALID_IPV6_TUNNEL_NETWORK (0X00001727). Happens most often when you're trying to add an IPv6 tunnel endpoint to a cluster network that doesn't have a proper IPv6 subnet assigned. I ran into this last month with a client who was setting up a Hyper-V live migration over an IPv6-only network. The cluster network was there, but the IPv6 prefix wasn't configured right.
Root Cause
The cluster checks that the network you're assigning the tunnel resource to has a valid IPv6 subnet. If the network doesn't have an IPv6 address range in the cluster configuration, or if the subnet mask is wrong, you get this error. It's not a driver issue. It's not a hardware issue. It's almost always a configuration miss in the cluster network settings.
The Fix
- Open Failover Cluster Manager – connect to your cluster.
- Go to Networks – in the left tree, expand the cluster and click Networks.
- Find the network that you want to use for the IPv6 tunnel. It's usually the same one you'd use for cluster communication.
- Right-click it and select Properties.
- Check the subnet – in the Properties window, look at the IP Addresses tab. You should see a valid IPv6 subnet listed, like
2001:db8::/64. If all you see is an IPv4 subnet, that's your problem. - Add an IPv6 subnet – if the network doesn't have an IPv6 range, you can't use it for an IPv6 tunnel resource. You have two options:
- If your physical network supports IPv6, configure an IPv6 address on the cluster nodes' NICs that belong to the same subnet. Then run the validation wizard again to let the cluster discover the subnet.
- If you're only using IPv6 for the tunnel itself and don't have a real IPv6 network, you can't do this with a regular cluster network. You need to create a separate network for the tunnel resource that's not tied to a physical subnet. That's an advanced scenario.
- Re-run the Validation Wizard – after the IPv6 subnet is visible in the network properties, try creating the IPv6 tunnel resource again. Should work.
If It Still Fails
Check these things:
- Duplicate IPv6 subnets – ensure you don't have two cluster networks with the same IPv6 prefix. That confuses the cluster and throws this error.
- NIC teaming – if the cluster network uses a teamed NIC, make sure the IPv6 address is bound to the team interface, not the individual adapters.
- Cluster service account – make sure the account running the Cluster service has permission to create network objects. Rare, but seen it with domain lockdowns.
- Event log – check the System and Cluster logs in Event Viewer for anything related to network validation. Sometimes you'll see a hint like "no IPv6 prefixes found."
If you still hit the wall, toss the network out of cluster configuration and re-add it. I've had to do that twice in the last year. Remove the network from the cluster (not the adapter), restart the Cluster service on all nodes, then let it rediscover the network.