Connect to ESXi via SSH
List all datastores to see the UUID of the problematic Datastore
# esxcli storage filesystem list
List active files in the datastore (the first column shows the required PID)
# lsof | grep '/ vmfs / volumes / 58580b99-a196935b-fa23-fc15b41d8300'
Replace 58580b99-a196935b-fa23-fc15b41d8300 with the UUID of Your problematic datastore
Kill each of the active PIDs in the failed datastore (They prevent You from unmounting the problematic datastore)
# kill 33172
Replace 33172 with the PID of Your active PIDs
Execute the command to unmount the datastore
# esxcli storage filesystem unmount -u 58580b99-a196935b-fa23-fc15b41d8300
Replace 58580b99-a196935b-fa23-fc15b41d8300 with the UUID of Your previously problematic datastore
Connect to ESXi and check the datastore, if necessary, mount datastore
Done
VMware errors:
Call HostStorageSystem.UnmountVmfsVolume for object failed
Operation failed, diagnostics report: Unable to find volume uuid