vCenter Server operations fail after upgrading to vCenter Server 5.0

Products

• VMware vCenter Server

Issue/Introduction

Symptoms:
After upgrading to vCenter Server 5.0, you may experience these issues:

• Editing virtual machine settings fails
• vMotion of a virtual machine fails
• In the vpxd.log you see errors similar to:
  
  info 'Default'] [VpxLRO] -- FINISH task-internal-5255 -- host-126984 -- VpxdInvtHostSyncHostLRO.Synchronize --
  warning 'Default' opID=AD23556A-00000086-87-4] [VpxdMoVm::PowerOn] Failed to relocate VM /vpx/vm/#219105/ to target host host-658, for reason : Not initialized
  error 'Default' opID=AD23556A-00000086-87-4] [VpxdMoVm::PowerOnInt] PowerOnIntImpl failed on VM /vpx/vm/#219105/ for unexpected reason : Not initialized
  error 'Default' opID=AD23556A-00000086-87-4] Alert:false@ d:/build/ob/bora-455964/bora/vpx/vpxd/vm/moVm.cpp:7560
  
  Note: The vMotion operation may succeed if you are logged in to vCenter Server using the vCenter Server web client.
  
  
• Creating a new virtual machine fails
• Virtual machine power on operation is failing with error:
  
  The general system error occured. Unexpected exception from PowerOnIntImpl
  
  
• You see an error similar to:
  
  A general system error occurred: Not initialized
  Call "EnvironmentBrowser.QueryConfigTarget" for object "envbrowser-xxx" on vCenter Server "xxxxxxxx" failed.

Environment

VMware vCenter Server 5.0.x

Cause

This issue may occur if stale entry of an inaccessible or inactive datastore on the ESX/ESXi host resides in the vCenter Server database. The inaccessible datastore may have been present before the upgrade to vCenter Server 5.0.

Resolution

This issue is resolved in VMware vCenter Server 5.0 Update 1. To download vCenter Server 5.0 Update 1, see the VMware Download Center.

If you are unable to upgrade, see the workaround below.

To work around this issue identify the inactive/inaccessible datastore that appears as grayed out in the vSphere Client connected to the ESX/ESXi host or vCenter Server.

• If it is an NFS datastore, right-click on the datastore and choose Unmount to remove the stale entry from the vCenter Server database.
• If it is an iSCSI or FC datastore, recreate the datastore with same parameters (LUN ID and datastore name) and present it to the ESX/ESXi host. When ESXi has access to it, you can migrate the virtual machines.
  

If the issue persists, perform one of these options:

• Ensure you do not have an inactive/disconnected host during the upgrade of vCenter and whether it is still reflecting in the vCenter inventory.
• Disable HA at the cluster level and migrate the virtual machines to other hosts and reboot the host. For more information on disabling HA, see Disabling VMware High Availability (HA) (1008025).
  
• Remove and re-add the ESX/ESXi host encountering the vCenter Server operation issues from the vCenter Server inventory.

Notes:

• If you cannot vmotion the virtual machines then you will have to shutdown the virtual machines then put the host into maintenance mode.
• If you remove and re-add the host, you lose performance data and resource pools.

• Reboot the ESX/ESXi host. If you reboot the host, there is downtime for the virtual machines as they cannot be migrated.

Additional Information

To be alerted when this document is updated, click the Subscribe to Article link in the Actions box Disabling VMware High Availability (HA)
vCenter Server 5.0 にアップグレード後 vCenter Server operations が fail する