vMotion Fails with Timeout Error After Moving Hosts to New vCenter/VDS
search cancel

vMotion Fails with Timeout Error After Moving Hosts to New vCenter/VDS

book

Article ID: 393785

calendar_today

Updated On: 04-10-2025

Products

VMware vCenter Server

Issue/Introduction

When attempting to vMotion virtual machines between ESXi hosts that have been recently migrated to a new vCenter Server and virtual distributed switch (VDS), the operation fails with timeout errors. The following error messages appear in the logs:

Migration to host <IP_ADDRESS> failed with error Timeout (195887137).
vMotion migration failed to receive data from the remote host: Timeout Failed waiting for data. Error 195887137.
vMotion migration failed to send buffer to remote host: Timeout.
vMotion migration failed writing stream completion: Timeout

These failures prevent successful migration of VMs between hosts in the same cluster, blocking normal operations such as host maintenance and upgrades.

Environment

  • ESXi 7.0 hosts being migrated to a new environment
  • vCenter Server 8.0
  • Virtual Distributed Switch (VDS)
  • Mixed environment where some hosts have been migrated successfully while others experience issues
  • Previous environment had NSX deployed, while the new environment does not
  • There are no MTU issues
  • Port 902 is open fine between the affected ESXi hosts in the cluster
  • Routing issues have been ruled out, such as, all affected host vMotion IPs are in the same subnet

Cause

The root cause is corruption or incomplete migration of VMkernel (vmk) interface configurations used for vMotion traffic. When ESXi hosts are moved from one vCenter Server to another, especially when moving from an NSX environment to a non-NSX environment, the VMkernel interface data in the vCenter database may not migrate correctly. This creates a situation where:

  1. The VMkernel interfaces appear correctly configured in the UI
  2. The hosts can see each other in the cluster
  3. However, the vMotion traffic cannot be properly established between some or all hosts

The database corruption affects how the hosts communicate for vMotion traffic, even though standard management traffic works correctly.

Resolution

  1. To resolve this issue, recreate the vMotion VMkernel interfaces on the affected hosts:

    1. Log in to the vSphere Client and connect to your vCenter Server.
    2. Navigate to the ESXi host experiencing vMotion issues.
    3. Go to Configure > Networking > VMkernel adapters.
    4. Create a new VMkernel adapter for vMotion:
      1. Click Add Networking.
      2. Select VMkernel Network Adapter and click Next.
      3. Select the appropriate distributed port group for vMotion traffic and click Next.
      4. Configure the appropriate IP settings (use the same IP scheme as your original vMotion network).
      5. Ensure the vMotion service is enabled for this VMkernel.
      6. Complete the wizard to create the new VMkernel adapter.
    5. Verify the new VMkernel adapter is properly configured with vMotion enabled.
    6. Delete the original vMotion VMkernel adapter that was migrated from the previous environment:
      1. Select the original vMotion VMkernel adapter.
      2. Click Remove and confirm the deletion.
    7. Repeat this process for all hosts experiencing vMotion issues in the cluster.
    8. Test vMotion functionality between the affected hosts to verify the issue is resolved.

Additional Information

  • Consider performing this solution proactively on all hosts being migrated if you encounter this issue with the first few hosts.
  • If the issue persists after recreating the VMkernel interfaces, re-check for other network configuration issues such as:
    • Physical network connectivity
    • MTU size mismatches
    • Firewall rules blocking vMotion traffic
    • VLAN configuration issues