Entering maintenance mode fails due to virtual machine auto migration failure

Products

• VMware vSphere ESXi
• VMware vCenter Server

Issue/Introduction

Entering maintenance mode fails under following conditions:

• DRS is enabled
• Slot policy option is enabled
• journal log shows following auto migration failure entries

MM DD HH:MM:SS <VC FQDN> vpxd[24601]: Event [1600315] [1-1] [YYYY-MM-DDTHH:MM:SS.89281Z] [vim.event.TaskEvent] [info] [VSPHERE.LOCAL\Administrator] [<DC name>] [1600315] [Task: Enter maintenance mode]
MM DD HH:MM:SS <VC FQDN> vpxd[24601]: Event [1600316] [1-1] [YYYY-MM-DDTHH:MM:SS.903249Z] [vim.event.EnteringMaintenanceModeEvent] [info] [VSPHERE.LOCAL\Administrator] [<DC name>] [1600315] [Host <ESXi FQDN> in <DC name> has started to enter maintenance mode]
MM DD HH:MM:SS <VC FQDN> vpxd[24601]: Event [1600317] [1-1] [YYYY-MM-DDTHH:MM:SS.987569Z] [vim.event.NoMaintenanceModeDrsRecommendationForVM] [info] [VSPHERE.LOCAL\Administrator] [<DC name>] [1600315] [Unable to automatically migrate <VM name#01> from <ESXi FQDN>]
MM DD HH:MM:SS <VC FQDN> vpxd[24601]: Event [1600318] [1-1] [YYYY-MM-DDTHH:MM:SS.987765Z] [vim.event.NoMaintenanceModeDrsRecommendationForVM] [info] [VSPHERE.LOCAL\Administrator] [<DC name>] [1600315] [Unable to automatically migrate <VM name#02> from <ESXi FQDN>]
MM DD HH:MM:SS <VC FQDN> vpxd[24601]: Event [1600319] [1-1] [YYYY-MM-DDTHH:MM:SS.987814Z] [vim.event.NoMaintenanceModeDrsRecommendationForVM] [info] [VSPHERE.LOCAL\Administrator] [<DC name>] [1600315] [Unable to automatically migrate <VM name#03> from <ESXi FQDN>]

…

• vpxd log shows "Required spare slots" is greater than "available slots" as below

YYYY-MM-DDTHH:MM:SS.903Z info vpxd[24882] [Originator@6876 sub=vpxLro opID=########-#####-auto-####-h5:########-##] [VpxLRO] -- BEGIN task-##### -- host-##### -- vim.HostSystem.enterMaintenanceMode -- ########-####-####-####-############(########-####-####-####-############)
…
YYYY-MM-DDTHH:MM:SS.987Z warning vpxd[24882] [Originator@6876 sub=DAS opID=########-#####-auto-####-h5:########-##] Required spare slots (38) > available slots (27)

Environment

VMware vCenter Server 7.x

VMware vCenter Server 8.x

VMware vSphere ESXi 7.x

VMware vSphere ESXi 8.x

Cause

As vpxd.log indicates, this issue occurs because the required slot for migration is larger than the available slot. 

Resolution

To resolve the issue, you need to reduce the slot size. Check the virtual machine settings in cluster to see if any resource reservations are configured, and if necessary, lower the reservation value or set it to zero (0).

To confirm slot size value via vSphere Client:  Select cluster -> Monitor -> vSphere HA -> Summary -> Advanced Runtime info 

 

Additional Information

Slot Policy Admission Control
https://techdocs.broadcom.com/us/en/vmware-cis/vsphere/vsphere/8-0/vsphere-availability/creating-and-using-vsphere-ha-clusters/vsphere-ha-admission-control/slot-policy-admission-control.html