Case # 04404492 Virtual Lab question

[TedLaurent]

I have a client who has a Hyper-V server and 2 VMWare servers. We made several attempts to migrate the Hyper-V Exchange VM but failed. I'm currently working on a migration project and part of that is Veeam and Virtual Labs. I have 2 virtual labs - 1 for Hyper-V and 1 for VMWare. I can run both virtual labs but they can't talk to each other so that puts limits on what I can test.

The Veeam server is a VMware VM that runs Hyper-v nested. The repository is local.

The VMware Virtual Lab is on the same host that the Veeam server runs on.

I thought that if I presented the VMware Virtual Lab Isolated VM Network to the veeam server and then pointed the Hyper-V Virtual Lab Isolated VMware Network to the new NIC that I would be able to have both labs communicate with each other.

I have tested the following:
- Ping VMware Isolated Lab appliance ( 192.168.31.1) from within the VMware Surebackup job - success
- Ping the Hyper-V static IP address of the NIC ( 192.168.31.140 ) from within the Veeam VMware Surebackup job - success
- Ping the Hyper-V Isolated Lab appliance ( 192.168.31.2 ) from within the VMware Surebackup job - fail
- Ping Hyper-V Isolated Lab appliance (192.168.32.2 ) from within the Hyper Surebackup job - success
- Ping the Hyper-V static IP address of the NIC ( 192.168.31.140 ) from within the Veeam VMware Surebackup job - success
- Ping the Hyper-V Isolated Lab appliance ( 192.168.31.1 ) from within the Hyper-V Surebackup job - fail
- Ping a VM in the VMware Surebackup job ( 192.168.31.229 ) from within the Hyper-V Surebackup job - fail
- Ping a VM in the Hyper-V Surebackup job (192.168.31.19 ) from within the VMware Surebackup job - fail.

While I have opened up a support ticket on this I'm assuming this is not a supported configuration. I know that some parts of this will fail if the virtual labs get reconfigured.

If you have any ideas on this I'd greatly appreciate hearing them.

Thanks,

Ted

[HannesK]

Hello,

The Veeam server is a VMware VM that runs Hyper-v nested.

typo, or really your setup? You run the Veeam server in the management partition of a Hyper-V host that is running in VMware?

After reading your ping tests a couple of time, I still do not know, which VM is where.

What also comes into my mind: Virtual switches in nested environments behave different behavior than normal switches from a MAC address learning perspective.

What is your goal? If you want different backups to talk to each other, what about just doing instant recovery to VMware in one separated network? Virtual lab is really not built to connect different labs.

Best regards,
Hannes

PS: I asked support to close the case... that kind of setup is job for support

[Andreas Neufert]

I know that issue from my labs. The issue is that you have virtual networks within a VMware VM.
You need to play with the ESXi vSwitch - Security - Policy Exceptions. Maybe enable all 3 and see if you can communicate and then disable one by one until you indetify the one that helps.

[HannesK]

"Promiscuous mode" is the setting I was pointing to (and Andreas probably means) when saying "different behavior than normal switches from a MAC address learning perspective."

[TedLaurent]

Thanks for the replies. David Domask with Veeam Support got me going in the right direction. Created virtual lab based on VMware backup. Do an Instant Restore of the Hyper-V VM into that virtual lab by mapping production network to virtual lab network (blows my mind that you can do this). Install VMware tools and set IP address. Viola! One big happy virtual lab.

[TedLaurent]

"Promiscuous mode" is the setting I was pointing to (and Andreas probably means) when saying "different behavior than normal switches from a MAC address learning perspective."

I do think that is the answer to the original issue. However the challenge with doing the jury rigging I had to do on the nested Hyper-V host is that any changes made the virtual lab overwrites the Hyper-V networking change that manual is done to point to appliances internal NIC to the VMware isolated network. Using Instant Restore I think is a much cleaner approach. Thanks for the post.

[TedLaurent]

typo, or really your setup? You run the Veeam server in the management partition of a Hyper-V host that is running in VMware?

Actually its not typo. It works rather well in fact. Thanks for the post.