vSphere Distributed Switch Health Check alarms not triggered when physical MTU is higher than vSphere Distributed Switch MTU

Products

• VMware vCenter Server
• VMware vSphere ESXi

Issue/Introduction

Symptoms

• vDS Health Check for MTU/VLAN is enabled.
• MTU mismatch is confirmed between ESXi hosts (e.g., 9000) and physical switches (e.g., 9216).
• vCenter Server does not trigger any health alerts or alarms.
• The Health Check status remains "Normal" or "Healthy."

Environment

VMware ESXi

VMware vCenter Server

Cause

This is the expected behavior of the Health Check feature.
The vSphere Distributed Switch Health Check uses the L2 echo-type protocol packets to probe the data path.
An alarm ("Not Supported") only triggers when there is an active risk of packet loss.
If the physical switch MTU is larger than the vSphere Distributed Switch MTU, packets originating from the vSphere Distributed Switch fit within the physical switch's limit, no packets are dropped, and the path remains healthy.

Resolution

No configuration changes are required. Maintaining a physical switch MTU slightly larger than the vDS MTU is an industry best practice to allow overhead for VLAN tags and network headers.

To verify the alarm system is functioning:

1. Lower the MTU on a single upstream physical switch port to 1500.
2. Wait approximately 5 minutes for the next layer 2 polling cycle.
3. Confirm a Not Supported alarm triggers in the vSphere Client.
4. Revert the physical switch port MTU to the original higher value (e.g., 9216) to clear the alarm.

Additional Information

For additional information, refer : vDS Health Check reports unsupported VLANs for MTU and VLAN