First off I would like to mention that i got this error:
Error: File <unspecified filename> is larger than the maximum size supported by datastore '<unspecified datastore>
After some searching i found out it was elated to blocksize, so i:
-deletet the failed backup
- went to my windows server where the backup repository is located and reformated the drive with 64K allocation unit.
From Veeam B&R i also click properties on the windwos server where the backup repository is located and followed the steps so it did "Collectin disk and volume info"..
Dont know if it was needed or not.

Run the job again, and it then failed with this error "Processing 'Servername'" and with the VM status "Failed".

Did'nt work.
The other message was gone, so the reformat might have fixed the the "Maximum size supportet by datastore" problem.
But it is only on large VMs (2TB+) I have gotten this error. Smaller VMs back up just fine to the same repository. So I am not sure what this message means.
The configuration is like this.
Production ESXi 5 has a datastore of like 8,8 TB.
The VM i try to backup has one 2 TB virtualdisk(Thin provsion), and one 250 GB (Thin provision). (only 53,7 GB used on that VM)
The backup repository is located on a windwos 2008 R2 VM on another ESXi 5 host.
This VM has 12 virtual disks each of 2TB. Windows 2008 R2 then span them to 1 volume of 24 TB.
The 24 TB volume is was first formated with NTFs and default allocation unit.
This is the volume that I reformated with 64K size.
Have looked in the logs, but did not find much. Any ideas?