DELL PPDM LWD based backups are failing due to error "Cannot complete the operation. See the event log for details. A disk was closed during the operation. This can happen due to VM lifecycle events such as VM power off, vMotion, and HA failover."

Products

• VMware vCenter Server

Issue/Introduction

• Performing LWD-based snapshot sync fails with the error "Cannot complete the operation. See the event log for details. A disk was closed during the operation. This can happen due to VM lifecycle events such as VM power off, vMotion, and HA failover."
• LWD Sync fails, followed by migration tasks on the vCenter.

Cause

The delta sync fails with a DiskClosing fault because of the DRM-recommended migration as seen in the logs below:

• In /var/run/log/dpd.log file of ESXi host, there are entries similar to:

YYYY-MM-DDTHH:MM:SS info dpd[497634223] [Originator@6876 sub=TransportIpc opID=24####24] Transport completed sync ####-##-###-##-#######-# for PE ####-##-###-##-#######-# with status The operation was canceled.
YYYY-MM-DDTHH:MM:SS info dpd[497634223] [Originator@6876 sub=TransportAdapter opID=24####24] Calling transport operation normal completion for Sync.####-##-###-##-#######-# with status The operation was canceled.
YYYY-MM-DDTHH:MM:SS error dpd[497634223] [Originator@6876 sub=Task opID=24####24] Task 'Dp.Task.Sync.####-##-###-##-#######-#' failed with error: (dp.fault.DiskClosing) {
--> faultCause = (vmodl.MethodFault) null,
--> faultMessage = <unset>
--> msg = ""
--> }

• In /var/log/vmware/vpxd.log file of vCenter server, there are entries similar to:

YYYY-MM-DDTHH:MM:SS info vpxd[08526] [Originator@6876 sub=vpxLro opID=lro--398369590-########-01] [VpxLRO] -- BEGIN task-66###07 -- vmname-- Drm.ExecuteVMotionLRO --
YYYY-MM-DDTHH:MM:SS info vpxd[08526] [Originator@6876 sub=drsExec opID=lro--398369590-########-01] Executing DRM recommended vMotion, migrating [vim.VirtualMachine:vm-#####,vmname] from vim.HostSystem:host-##### to vim.HostSystem:host-#####
YYYY-MM-DDTHH:MM:SS info vpxd[08526] [Originator@6876 sub=VmProv opID=lro--398369590-########-01] Determined workflow protocol version of 3

Resolution

This is expected behavior of the DRS load balancer.

1. Adjust "Migration Threshold" on the DRS from vCenter UI -> Cluster -> Configure -> vSphere DRS -> EDIT -> Migration Threshold.
2. Sliding the "Migration Threshold" left reduces the vMotion aggressiveness.
3. Change the threshold level from 3 to 2 and click Ok.
4. This should increase the tolerance capacity of the DRS and migrate VMs less frequently.

Additional Information

Dell KB suggests restarting TSDM and DPD services on the ESXi host, which is a temporary fix.

Ref: https://www.dell.com/support/kbdoc/en-us/000217246/ppdm-vm-backup-failed-with-code-avb0016