Backup failed. SCSI device mount failed. Lun:0, Uuid: xxxx

[mmeier]

Hi,

getting this error when trying to backup AHV VMs on new Nutanix cluster:
Backup failed. SCSI device mount failed. Lun:0, Uuid: xxxx

Environment:
Nutanix AHV cluster with 3 nodes on version 5.15.2 (LTS)
Veeam B&R 10a
AHV Proxy: 2.0.558
VUL perpetual license

Seems to be, that there is something going wrong on step:
"AHV Backup Proxy on the AHV cluster creates a volume group (VG) and mounts VM disks via iSCSI and retrieves VM data on block level."
In Prism, we can see, that a volume group was created.

Any ideas how to lock / fix?

Best Regards,
Matthias

[ronnmartin61]

Hey Matthias, yes this is precisely how the backup process for AHV works. This is likely a tech support proposition but one thing you may want to check is whether or not the VM failing has a disk, network or some other VM attribute that no longer exists as we have seen some odd side effects in these cases.

[mmeier]

Hi Ron, thank you for the quick reply.
I already opened a tech support case.
Seems not, that the VMs have missing elements. VMs were created from scratch.

[mmeier]

Problem is solved. There was a typo within the Nutanix cluster config regarding the “iSCSI Data Services IP”. Therefore, the mount was failing.

[otrobison20]

Hi
I am seeing the same issue, the typo was it a IP typo or Nutanix config typo ?

[PetrM]

Hi Oli,

Please reach out our support team and don't forget to share a support case ID with us.

Thanks!

[otrobison20]

Hi thanks for you reply I already have a case open but dont seem to be getting far #04794464

[PetrM]

The case has been created yesterday and our engineers already work on it, please let them have some time to figure out the root cause.

Thanks!

[ibarizz]

Hello,

I am having a similar scenario to this, yet we can't find a solution to this.
Also support is not helping us in the way we wanted. Case support is #06296031.
If somebody can help me, I'd be appreciated.

Cheers.

[PetrM]

Hello,

Please don't hesitate to request an escalation of your support case if you feel that the investigation goes sideways.

Thanks!

[ibarizz]

Thank you @PetrM
We managed to solve it by my own. It looked like the iSCSI Data Services IP was poing to a load balancer of a few CVMs that were decomissed during a migration of our Nutanix cluster' nodes.
Ye, it has to be always a networking problem!