Error: Cannot complete file creation operation.

[unsichtbarre]

Case # 04776374

Versions: vSphere 7 / vCenter 7 / ESXi 7 /vSAN 7 / Veeam 11.0.0.837
vSAN: 4 Node 150TB RAW, vSAN Default Storage Policy (unaltered) RAID 1 / 75TB Available

When I try to seed a replication job to the vSAN, it fails with the error: Error: Cannot complete file creation operation. The 2 VMs I am trying to seed are 50GB each

VMware validates storage policies and vSAN configuration as correct, Veeam support (at least my representative) is not willing to consider the issue.

One other thing, https://kb.vmware.com/s/article/2105091 is not the issue due to my version (7.X) and HardwareAcceleratedLocking is already enabled.

Thanks for consideration,
-JB

[HannesK]

Hello,

Thanks for consideration,

please consider to continue working with support, as guessing issues like this makes little sense on a forum.

As far as I understand the case, the logs say that the issue comes from the VMware side (which might be based on similar issues, where it was a VMware issue). I also see a note, that permissions were an issue in the first place.

I also assume that you run vSphere 7.0 or 7.0U1 and not U2 (U2 is not supported yet)

Best regards,
Hannes

[unsichtbarre]

I will continue to work with support, however the agent indicated that Veeam support considered this to be a VMware problem. VMware considers vSAN to be correct, and so (therefore) not their problem. Also, no errors appear in VMware or VMware logs, only Veeam logs. Yes vSphere/vSAN 7.0.

I consider the permission issue to be related to full admin, the error changes but does not disappear. Interestingly, if the job is seeded to a normal SAN datastore, then I migrate the VM to vSAN, it will continue to replicate without issue. It is only when I seed to vSAN that the issue happens.

I can seed to normal SAN with the default permissions.

THX,
-John

[PetrM]

Hi John,

I'm not sure that you really should reach out VMware support at this stage of investigation, the issue must be confirmed by precise analysis of VMware logs and our logs. One more idea would be to define the exact operation which is failed and to try to reproduce the same problem without Veeam, for instance by invoking the API call manually. I've just asked our support team to review the case and to try to help you to isolate the issue at vSphere level and to communicate with VMware support. Also, you may request an escalation of this support case.

Nevertheless, I'm pretty sure that this issue does not come from our code, perhaps this article and that blog might give a better idea, you'll find different examples of failed attempts to create vSAN objects.

Thanks!

[unsichtbarre]

Thanks for the pointers: verified MTU, verified CLOMOD Liveliness

Still no folder creation. It is a VMware issue, but no resolution so far

[unsichtbarre]

The problem seems to be that Veeam is not selecting a storage policy. As I set the target Datastore, I see the vSAN datastore listed under both "Default Policy" (where the vSAN is listed alongside other IP SAN datastores) and in a separate category "vSAN Default Storage Policy". I have tried selecting both, to no effect, it always fails with: Error: Cannot complete file creation operation.

VMware says Veeam is simply not choosing a Storage Policy (even though it looks like I pick one) and, therefore, the files cannot be created.

[PetrM]

Hi John,

Ok, thanks for information. I have 2 questions:
1. How did VMware engineer come to the conclusion that the issue is related to storage policy selection?
2. Have you tried administrator account as it was suggested by our engineer?

Also, please share this information with our support as well, it's a very important note.

Thanks!

[unsichtbarre]

Yes we tried Administrator.

Issue seems to be that the vSAN default policy (set on Datastores > vSANVolume > Configure -- https://docs.vmware.com/en/VMware-vSphe ... 4C670.html ) needs to match the policy selected in the replication job. The folder seems to be created before the replica storage policy is applied.

Resolved now. THX for all the help

[PetrM]

Hi John,

Glad to hear that the issue is resolved and thanks for sharing the result with the Community!

Thanks!