vMotion fails after network configuration changes
search cancel

vMotion fails after network configuration changes

book

Article ID: 451362

calendar_today

Updated On:

Products

VMware vCenter Server

Issue/Introduction

vMotion migration attempts may fail after a vCenter server patch, upgrade, or hardware maintenance. Symptoms include "Host is down" errors in migration logs and total loss of connectivity over the vMotion VMkernel adapter, despite host-level configurations being correct. This article addresses issues stemming from environmental network changes.

Environment

  •  vSphere 7.0, 8.0, 9.0
  • VMware Cloud Foundation (VCF) 4.x, 5.x.

Cause

This issue is typically environmental and not caused by the software upgrade itself. Common root causes include:

  • VLANs used for vMotion (e.g., VLAN 404) not being properly tagged or allowed on physical switch trunk ports.
  • Physical switch configuration changes (e.g., port pruning or access control lists) during maintenance.
  • Inconsistent load-balancing or teaming policies across the cluster.

Resolution

  1. Verify connectivity using vmkping: Run the following command to test vMotion network reachability: vmkping -I vmk1 <destination_IP>
  2. Inspect VLAN configuration: Check if the required VLAN is tagged and allowed on the physical switch ports connected to the ESXi host uplinks.
  3. Utilize CDP/LLDP info: Navigate to the ESXi host in vCenter (Configure > Networking > Physical Adapters) and retrieve the CDP/LLDP information. Provide this data to your networking administration team to verify that the upstream physical switch ports are correctly configured to trunk the vMotion VLAN.
  4. Verify ESXi Firewall: Ensure that the "vMotion" service is enabled under Configure > System > Firewall for all hosts involved in the migration.
  5. Review physical switch logs: Ask the network team to review physical switch logs for dropped packets or interface errors corresponding to the time of the vMotion failure.

Additional Information