vMotion compatibility alert: "Too large clock skew was detected" on time-synced ESXi hosts

Products

• VMware vSphere ESXi
• VMware vCenter Server

Issue/Introduction

• When attempting to perform a vMotion, the migration wizard displays the following compatibility alert: Too large clock skew was detected. Relative time skew ######## between source and destination hosts is greater than five minutes

• vMotion operations continue to function and complete successfully despite the warning message.

• NTP is configured correctly on both source and destination ESXi hosts, and no time difference is observed in the vCenter UI. This can be verified in vCenter UI by navigating to ESXi host > Configure > System > Time Configuration

• Following entries are recorded in the /var/run/log/vpxa.log on the affected ESXi host:

YYYY-MM-DDTHH:MM:SS In(###) Vpxa[#######]: [Originator@#### sub=vpxaMoService opID=###########################] Time difference between vpxd and vpxa is ### seconds

Environment

VMware vCenter Server 7.x
VMware vCenter Server 8.0 Update 3e and below.

Cause

This issue occurs because of a stale time-drift status in vCenter. It occurs when a previously detected clock skew is resolved while the ESXi host's VPXA service is in a stopped state. As the service was inactive during the time synchronization, it could not communicate the update to the vCenter VPXD service. As a result, vCenter continues to incorrectly report that a clock skew exists.

Resolution

This issue is fixed in vCenter 8.0 Update 3g. Log in to the Broadcom Support Portal to download the patch. 

Workaround :

 Restart the VPXD service on the vCenter Server Appliance to clear the stale alert.

1. Log in to the vCenter Server Appliance (VCSA) via SSH.

2. Restart the vpxd service using the below command:
   service-control --restart vpxd

Additional Information

Too large clock skew was detected