Failed to modify disks when doing an Instant restore to HyperV

[Taroking1]

We are using Veeam to do our Virtual to Virtual (V2V) migrations from ESX to HyperV. We backup up from ESX and restore to HyperV using Instant Recovery with no agent installed. This functionality is absolutely fantastic!!

We came across a problem recently that I wanted to share in case anyone else has the same issue.

Scenario
Back up Linux VM from ESX
Perform an Instant Recovery on the Linux VM to HyperV host
During this Instant Recovery a temporary helper VM is used that has it's own IP address. We are using a fixed IP address for this.
We have used this functionality successfully at a few sites already

Problem
When doing a restore to one particular site, we kept getting errors about unable to modify disks.
NOTE: This problem is normally caused by missing firewall rules and blocked traffic. This should be the first thing to double check and is likely to be the cause of this error if you see it.
We checked all of our firewall rules and they were all correct. We lodged a call with Veeam and they also confirmed that our rules were all in place and correct.

Investigation
As per Veeam: The config.vhdx file contains network configuration.If a path is set inside the CSV, then the mount is unsupported and the operation getting failing.
So this was our actual issue. When doing an instant Recovery we are indeed pointing the recovery to a CSV disk

Workaround
If we come across this issue, we restore the VM to a local disk on the HyperV server
After the recovery has completed we then just move the VM over onto the CSV and it is done. Adds 5 mins to the process, but still fairly painless while we wait for a fix in v12 maybe?

[HannesK]

Hello,
do you maybe have the case number where that was recommended to you? Because having local storage as a requirement depending on the network configuration sounds wrong to me (and I can remember I tested it successfully directly to a CSV)

Can you maybe also give some more details on the config.vhdx? I don't see that file for my Hyper-V VMs.

Thanks,
Hannes

[Taroking1]

Hi Hannes

Case #05076713

Having local storage is definitely not normally a requirement. We have done this exact same process successfully before and straight to CSV with no issues. This is the first site we have come to in our migration where we have experienced this.

[HannesK]

Hello,
sounds like issue #335172 which is related to GPT partitions.

The workaround is what support told you. If it is issue #335172 (support can confirm that), then there are plans to fix it in V12.

Best regards,
Hannes

[madshark]

Hi, I know this is an older post now, but wondering if there has been any updates as to whether this is fixed in V12?
I am running V12 and am seeing this when migrating Linux VMs from VMware to Hyper-V (Cluster with CSV's). The migrations do seem to be successful and VMs work fine after migration.
It seems something that can be ignored, but we want to understand it and be sure, if possible.

[HannesK]

Hello,
bug 335172 was fixed, yes.

If you see any issues, please open a support case and post the case number for reference.

Best regards,
Hannes