Management network outage during physical switch maintenance on Multiple ESXi hosts.

Products

• VMware vSphere ESXi

Issue/Introduction

• Multiple ESXi hosts become "Not Responding" in vCenter during physical switch maintenance or upgrades.
• Virtual machines resident on impacted hosts lose network connectivity.
• The issue affects multiple clusters simultaneously.
• Physical link status for redundant NICs shows as "Up," but no traffic passes.

Environment

• VMware vSphere ESXi 7.x / 8.x

Cause

1. The outage occurs because the Management Network port group (vmk0) is configured with an explicit failover order override where one physical NIC (e.g., vmnic0) is Active and the redundant NIC (e.g., vmnic3) is Unused.
2. When the physical switch connected to the active NIC is taken offline, the ESXi networking stack does not failover to the redundant NIC because it is explicitly prohibited by the "Unused" state.
3. To verify the configuration : Navigate to the Management Port from the Switch setting and verify the Teaming & Failover option :
   
   

Resolution

To restore management redundancy and prevent outages during maintenance, move the redundant uplinks to a Standby or Active state:

1. Log in to the vSphere Client.
2. Navigate to the affected ESXi host > Configure > Networking > Virtual switches.
3. Locate the virtual switch (e.g., vSwitch0) and expand the Management Network port group.
4. Click Edit Settings.
5. Select Teaming and failover.
6. Locate the redundant adapter (e.g., vmnic3) in the Unused adapters list.
7. Use the arrow icons to move the adapter into the Active adapters or Standby adapters list.
8. Click OK.
9. Repeat these steps for all hosts in the cluster to ensure consistent redundancy.

Additional Information

For more details on teaming policies, refer Configure-nic-teaming-and-load-balancing