ESXi host takes extended time to connect to vCenter after patching

Products

• VMware vCenter Server
• VMware vSphere ESXi 8.0

Issue/Introduction

Following an ESXi host reboot after applying patches using vSphere Lifecycle Manager (vLCM), you observe the following:

• The ESXi host takes significantly longer (several minutes) to show as Connected in the vCenter Server inventory compared to a standard reboot.
• The ESXi host boots to the DCUI screen quickly, but remains in a Not Responding or Disconnected state in vCenter.
• In the vCenter vpxd.log, you see events similar to: VPXA heartbeat build changed Triggering HostDisconnectAndReconnectLRO

Environment

• VMware vSphere ESXi 7.x
• VMware vSphere ESXi 8.x
• VMware vCenter Server 7.x / 8.x

Cause

This behavior is by-design. When an ESXi host is patched and its vpxa agent build number changes, the vCenter Server detects the build change through the heartbeat mechanism.

Unlike a standard reboot where the build remains the same (allowing for a lightweight resync), a build change triggers a full HostDisconnectAndReconnectLRO (Long Running Operation). This internal reconciliation process is required to:

• Re-authenticate the management session (SOAP/session re-validation).
• Reconfigure HA (FDM) agents.
• Re-execute cluster membership and EVC compatibility checks.
• Re-sync Trusted Infrastructure policies and re-assign licenses.

Resolution

This is expected behavior and no corrective action is required.

When planning maintenance windows using vSphere Lifecycle Manager, ensure that additional time is factored in for each host to complete the vCenter Server reconciliation process after the hardware/kernel boot is finished.