Prefered network not obeyed with agent?

[mdiver]

I observed the following with the most recent versions of the product (V11/V5).

Consider managed agent backup job. Prefered network is set to a certain backup network. The physical workload with the agent has a NIC in this network.
VBR has NICs in a mangement network and also in this backup network. For VM backups this is respected. For the agent backup we found that the initial control channel (TCP:10005) is always first driven to the management FQDN of the VBR. As this is not reachable it doesn't work. Once I set a host entry on the agent machine with the backup network IP behind the managment FQDN it starts working.
So it seems the agent isn't informed about the prefered network settings. Is this by design?

We found reg keys steering this. But they always revert back once we tamper with them:

Code: Select all

HKLM/SOFTWARE/Veeam/Veeam Endpoint Backup/BackupServerIPAddress mgmt-ip
HKLM/SOFTWARE/Veeam/Veeam Endpoint Backup/JobSettings mgmt-fqdn:10005
HKLM/SOFTWARE/Veeam/Veeam Endpoint Backup/VBRServerName mgmt-fqdn

Thanks,
Mike

[Dima P.]

Hi Mike,

We are investigating this issue but kindly ask you to raise a support case and share the case ID with us. We will need the VBR and agent logs to check how the network rules are applied and executed. Thank you in advance!

[mdiver]

@Dima P.
Thanks for your response. I'll follow up once I opened the case. It will take another week.

[mdiver]

Case was opened: #04855323

[Dima P.]

Passed to QA team. Thanks!

[mdiver]

Problem could not be solved by support. Case was therefore closed.
Customer used workaround with split-DNS (alike the host entry described above).

Maybe someone could check if this is by-design?

Thanks,
Mike

[Dima P.]

Mike,

Yes we are still checking the issue with QA team and will try to reproduce it based on your case details. Thanks for the update!