Windows failover cluster adapter media disconnected

Hi Booska,

That Cluster Virtual Adapter showing “Media disconnected” is normal in Hyper-V failover clustering, so you do not need to manually connect it. It’s just an internal cluster adapter, not a real network card. The actual issue is more likely that the two nodes are having trouble talking to each other over the network path used for cluster storage traffic, which is why you’re seeing SMB errors, storage paused warnings, and the first node slowing down when the second node joins. I’d check that the NIC used for cluster communication has both Client for Microsoft Networks and File and Printer Sharing enabled, and make sure that network is allowed for cluster communication in Failover Cluster Manager. If those are already correct, the next most common cause is the NIC driver or advanced NIC features, so try temporarily disabling things like Jumbo Frames, VMQ, or Large Send Offload and then rerun validation.

Harry.

The cluster virtual adapter (NetFT adapter) showing as “media disconnected” is expected and does not need to be manually “connected.” The real problem is the CSV/SMB path between nodes becoming unstable when the second node joins, which causes CSV I/O redirection issues and storage pauses.

Use these steps to troubleshoot and stabilize the cluster instead of focusing on the virtual adapter state:

1. Verify CSV/SMB connectivity and ports
   • Ensure all required cluster and SMB ports are open between nodes (especially TCP 445 for SMB and UDP/TCP 3343 for cluster traffic) on any firewalls between the nodes.
   • On each node, verify firewall rules for Failover Clustering and SMB, for example:

          netsh advfirewall firewall show rule name="Failover Clusters (UDP-In)" verbose
     

   • Confirm that ICMP (ping) is allowed between nodes for validation tests.
2. Check CSV network configuration
   • CSV traffic uses SMB between nodes. If the preferred cluster network for CSV becomes unavailable or is misconfigured, the cluster will fail over to another network that may not allow SMB, which produces Event ID 5120/5142 and CSV paused/queued I/O.
   • Ensure that any network the cluster might use (any network not disabled for cluster use) is configured to allow SMB/CSV traffic. If a network is enabled for cluster use but blocked for SMB, CSV redirection will fail.
   • Avoid changing cluster network priorities manually; let the cluster choose the preferred network based on metrics unless there is a specific, documented design.
3. Investigate CSV paused/storage warnings
   • The symptoms described (node sluggish, storage paused warnings, SMB communication errors) match common CSV paused scenarios:
     • Network bottleneck or misconfiguration (including NIC teaming mismatch across nodes).
     • Network adapter resource exhaustion.
     • Physical disk/HBA or storage fabric issues.
   • Recommended checks:
     • Align NIC teaming configuration on all nodes (Get-NetLbfoTeam) and ensure teaming mode and load-balancing are identical.
     • Consider dedicating a network (or VLAN) for CSV and Live Migration traffic to avoid congestion.
     • Check for Event IDs 5120/5142/153 on the nodes and investigate any underlying disk or HBA issues.
4. Review cluster networks and validation
   • Run a fresh Cluster Validation from Failover Cluster Manager with both nodes present and review the Network and Storage sections for warnings/errors.
   • Pay particular attention to:
     • Network tests that show partitioned networks or failed SMB connectivity between nodes.
     • Any warnings about network congestion or packet loss.
   • If validation reports partitioned networks or heartbeat issues, follow guidance to correct NIC teaming, driver, or switch configuration.
5. Check for NIC/driver and virtualization layer issues
   • Outdated NIC drivers or misconfigured virtual NICs (for example, in VMware or other hypervisors) can cause packet loss and node-to-node communication failures, which in turn cause CSV pauses and node regroup events.
   • Update NIC drivers and firmware per vendor guidance and verify switch configuration and cabling.
6. Re-test after fixes
   • After correcting network/teaming/firewall/storage issues, re-run Cluster Validation and confirm that SMB communication tests pass and that CSV no longer enters a paused state when the second node joins.

The cluster virtual adapter’s “media disconnected” status in ipconfig is not itself a fault condition; focus on ensuring that all cluster-enabled networks support SMB and that NIC teaming, drivers, and storage paths are correctly configured and healthy.

References:

• Troubleshoot storage issues in Hyper-V and Windows Server failover clusters
• Troubleshoot cluster issue with Event ID 1135
• Cluster Shared Volumes overview
• Unable to access ClusterStorage folder on a passive node in a server cluster