VMware Backups using NBD transport are failing with error "unexpected termination of client"

Problem

• Specific VM backup fails with status 6 with nbd transport mode.
• This issue can occur at any NetBackup version

Error Message

Job Detail Status:

Oct 5, 2022 12:13:47 PM - Error bpbrm (pid=13215) from client <vm_name>: ERR - Error opening the snapshot disks using given transport mode: nbd Status 23
Oct 5, 2022 12:13:48 PM - Critical bpbrm (pid=13215) from client <vm_name>: FTL - cleanup() failed, status 6
Oct 5, 2022 12:13:51 PM - Error bptm (pid=13286) media manager terminated by parent process
Oct 5, 2022 12:13:57 PM - Info <backup_host> (pid=13286) StorageServer=PureDisk:<backup_host>; Report=PDDO Stats for (<backup_host>): scanned: 2 KB, CR sent: 0 KB, 
CR sent over FC: 0 KB, dedup: 100.0%, cache disabled, where dedup space saving:100.0%, compression space saving:0.0%
Oct 5, 2022 12:13:57 PM - Critical bpbrm (pid=13215) unexpected termination of client <vm_name>
Oct 5, 2022 12:13:57 PM - Info bpbkar (pid=0) done. status: 6: the backup failed to back up the requested files
Oct 5, 2022 12:38:45 PM - Info nbjm (pid=5336) starting backup job (jobid=3665986) for client <vm_name>, policy <policy_name>, schedule 
FULL
Oct 5, 2022 12:38:45 PM - estimated 0 kbytes needed
Oct 5, 2022 12:38:45 PM - Info nbjm (pid=5336) started backup (backupid=<vm_name>_1664966325) job for client <vm_name>, policy <policy_name>, 
schedule DC1_VMWARE_FULL on storage unit stu_disk_<backup_host> using backup host <backup_host>
Oct 5, 2022 12:38:46 PM - started process bpbrm (pid=13215)
Oct 5, 2022 12:38:49 PM - connecting
Oct 5, 2022 12:38:49 PM - connected; connect time: 0:00:00
Oct 5, 2022 12:38:52 PM - begin writing
Oct 5, 2022 12:39:04 PM - end writing; write time: 0:00:12
User backup failed  (6)

vxms logs show error: Fail to get vm parent.

The vxms log location at VMware Backup Host Server

• Unix: /usr/openv/netbackup/logs/vxms
• Windows: install_path\NetBackup\logs\vxms

VixDiskLib: VixDiskLib_OpenEx: Open a disk.
VixDiskLib: Error occurred when obtaining NFC ticket for [<datastore_name>] <vm_name>/<vm_name>_6-000002.vmdk. Error 1 (Unknown error) 
(Fail to get vm parent) at 2547.
VixDiskLib: VixDiskLib_OpenEx: Cannot open disk [<datastore_name>] <vm_name>/<vm_name>_6-000002.vmdk. Error 1 (Unknown error) at 5769.
VixDiskLib: VixDiskLib_Open: Cannot open disk [<datastore_name>] <vm_name>/<vm_name>_6-000002.vmdk. Error 1 (Unknown error) at 5813.
VixDiskLib: VixDiskLib_Disconnect: Disconnect.

Cause

• VDDK might be having trouble finding the base vmdk as per the error: Fail to get VM parent
• Error is due to the VM's configuration about the parent not being set.
• It can be seen from the picture below that either the "parent" or "parentVApp" needs to be set for the VM.  If all of these are "Unset", this error will be seen.
• Below screenshot is from vCenter MOB  with the following URL that can be used on the browser where vCenter is accessible.  Note that use of the administrator@vsphere.local account credentials are required to login.

https://<vcenter_ip>/mob/?moid=vm-<id>

Solution

• This failure can be due to the location configuration of the Virtual Machine at vCenter. It would be under vCLS folder (group-v4 in above example) which is created automatically and only vCLS(cluster services) VMs should be a part of this folder. 
• By default, a back up the VMs part of this folder is not needed.
• After moving the VM back to a normal folder in vCenter console, backups should complete successfully.