vMotion fails due to inconsistent Uplink Teaming policy - VMware vSphere
Article ID: 453023
Updated On:
Products
Issue/Introduction
vMotion migrations frequently fail in stretched or multi-site clusters when network pathing becomes asymmetric.
The vSphere Client displays the error: The vMotion failed because the destination host did not receive data from the source host on the vMotion network. Please check your vMotion network settings and physical network configuration and ensure they are correct.
Connectivity verification shows unidirectional traffic: ICMP packets from Site 1 reach Site 2, but response traffic from Site 2 fails to return to Site 1.
vmkping from the vMotion adapter (e.g., vmk1) shows 100% packet loss between hosts in different sites.
The vmx log contains entries similar to: Migration [-1062699939:5181212996128058746] failed to connect to remote host <####> from host <####>: Host is down.
Environment
- VMware vSphere ESXi 8.x
- VMware vCenter Server 8.x
- Multi-site/Stretched Cluster environment.
Cause
Asymmetric network pathing caused by mismatched Active/Standby uplink order on vMotion port groups across different physical sites.
This issue is caused by inconsistent Uplink Teaming and Failover (Active/Standby) policy configurations on the Distributed Switch (vDS) or Standard Switch port groups across sites.
This leads to asymmetric network pathing where egress traffic from one site is dispatched via an unroutable or mismatched physical interface, preventing the bi-directional communication required for vMotion.
Resolution
To resolve this issue, ensure symmetric bi-directional network communication by aligning the uplink order:
- Log in to the vSphere Client.
- Navigate to Inventory > Networking.
- Select the Distributed Switch and go to the Configure tab.
- Select Settings > Topology to identify the uplinks assigned to the vMotion port group.
- Right-click the vMotion port group and select Edit Settings.
- Navigate to Teaming and failover.
- Verify the Failover order. Ensure the Active uplinks and Standby uplinks are identical across all hosts in the cluster, regardless of their physical site.
- If the hosts at Site 2 are using a different uplink order than Site 1, reconfigure them to match.
- Test connectivity using
vmkping -I vmkX <Destination_vMotion_IP>from both sites to confirm bi-directional traffic.
To speak with a customer representative or a Support Engineer see . Scroll to the bottom of the page and click on your respective region.
Additional Information
For instructions on retrieving ESXi logs to verify network timeouts, see .
Feedback
