Removed ESX host from vCenter server and unable to re-add to VC inventory with error "An error occurred while communicating with the remote host"

Products

• VMware vSphere ESXi

Issue/Introduction

• Large number of VMs managed by ESX host.
• Using vSphere Replication.
• ESX logs "/var/run/log/hostd.log" show errors.
  
  

  2026-01-29T13:51:42.330Z Wa(164) Hostd[2106361]: [Originator@6876 sub=Vmsvc.vm:/vmfs/volumes/<path>/<vm_name>.vmx] File - failed to get objectId, '/vmfs/volumes/<path>/<vm_name>.vmx': Operation not supported (11)

  2026-01-30T16:39:53.821Z Wa(164) Hostd[2099578]: [Originator@6876 sub=Vimsvc.HaSessionManager opID=45442b1b sid=52cea133] Rejected password for user from <ip_address> - session=52cea133-c8a0-5a34-fa72-56ecf95b9c2d

  2026-01-30T16:39:53.821Z In(166) Hostd[2099578]: [Originator@6876 sub=Vimsvc.ha-eventmgr opID=45442b1b sid=52cea133] Event 1858 : Cannot login @<ip_address>


  • Followed steps from KB Reconnecting an ESXi Host to vCenter Server fails with the error: Rejected password for user vpxuser and issue persists.
• ESX logs "/var/run/log/vpxa.log" show errors.
  
  2026-02-02T20:37:59.999Z Cr(162) Vpxa[2262123]: --> Panic: Unable to allocate memory

Cause

A scheduled job (e.g., Veeam backup solution, etc.) is creating snapshots daily of the VMs managed by the ESX host, without deleting (consolidating) the old snapshots.

• Eventually the number of snapshots increase to the point where the associated memory overhead exceeded the memory limit allocated to the vpxa service.

• Possibly due to the scheduling time of when backup solution (or other scheduled task) completes and when Replication tasks start.

  • It is possible the task is successfully deleting the snapshots after completion (e.g., backup job), but the Replication tasks are capturing the snapshots before they are deleted.

Resolution

Delete all VM snapshots on ESX host.

• If unable to delete the snapshots due to replication or other requirements, unregister the VMs from the ESX host.

Once the snapshots have been deleted for all VMs or VMs have been unregistered, re-add to the VC inventory.

Additional Information

• Work with backup vendor support or review whatever is scheduled to perform the snapshot creation tasks to determine why it is failing to delete the old snapshots. Resolve the issues found to make sure the old VM snapshots are deleted regularly.

• The scheduling of the backup/scheduled snapshot jobs and the replication tasks may need to be adjusted.