5.0 Active Directory Item Restore Process Issues

[analyst]

We have been unable to successfully use the AD Item Restore or any other Item restore process. I have some questions about the functionality and configuration of the Virtual Lab and the applicances.

I've created an Application Group which contains a DC which has been successfully backed up. Created the Virtual Lab and therein lies my question. I have concerns when configuring the Production and Isolated network Interfaces. I understand that you can create the Isolated network. When I do this however I am unable to Ping the gateway of the isolated network. Is this be design? Also, I do not feel comfortable with providing a Production interface that has connectivity to the isolated network when I'm powering on a DC. Primarily a DC which has the Infrastructure Master on it.

The documentation on this process is extremely poor in my opinion. I would like to know more about this entire process as it is not very clear at all. Is there any addition documentation besides the User Guide that tells what is going on under the hood and how this is working?

[Gostev]

Yes, I have pretty detailed explanations in sticky FAQ topic. That should should cover most questions you might have. Also, if you have any issues setting up virtual lab and using the product, please do not hesitate to contact our support for assistance.

You cannot ping gateway in the isolated network, because the network is... err... fully isolated from production, and thus you cannot reach your real gateway that is located on the production network. Why do you need to be able ping gateway in the isolated network? You should only need this when you have multiple isolated networks in your virtual lab (to mirror multiple isolated networks in production). In that case, as described right in the wizard, you need to give the proxy appliance the same IP address in each isolated network, as IP address of your gateway in the corresponding production network. This way, proxy appliance will act as gateway, and route traffic between those multiple isolated networks.

[analyst]

I'm sorry but I cannot find the documentation that you are referencing concerning your "sticky FAQ". The reason that I asked if I should be able to ping the Gateway was precisely because of the situation you described concerning multiple networks. I was unaware that it would only provide a gateway when creating multiple networks in the isolated environment. I assumed that a gateway would be created by default irregardless if a single or multiple subnets were being created. Now I know. Thank you.

The reason I was pinging the gateway was to trouble shoot errors that were generated on the failed recovery attempt. The errors stated that it was not able to ping the IP address of the server being resolved or something to that effect. Again...since the documentation does a very poor job of explaining how this process is actually working with the virtual applicance it makes it rather difficult for an end user to troubleshoot if they do not know how thie VA is working or configured.

I do not understand why a Production Interface is required if the entire environment is ...err..."isolated" unless I want to provide access from my Production network to the Isolated environment via a Static Mapping which I do not. "Isolated" to me suggests that it should have any dependancy on a Production environment.

Could you provide a link to this document? Thank you.

[Gostev]

It is right under New Topic button here in Veeam Backup & Replication forum I am definitely accepting suggestions on how this FAQ can be made more noticeable, if you have some ideas.

You want to see Virtual Lab section in FAQ to start with. Let me know if you still have any further questions after looking through it. Thanks!

[analyst]

I was able to log into the isolated DC to see what is happening. The problem is the Sure Backup Job machine is assigning a different adapter to the isolated DC with a different IP address. Therefore it cannot resolve DNS from itself or complete the verification process for the GC, DC or DNS scripts ran against it. What are we doing wrong?

[Gostev]

SureBackup job does not touch any network settings of any VM it powers on in the isolated environment. We do not have such logic, and there is simply no need to do this, because isolated environments is a complete copy of production environment.

Does this VM use Windows 2008 R2 and VMXNET3 adaptor, by any chance?

[analyst]

Yes

[Gostev]

OK, that's what I thought. There is an issue currently with how VMware Tools 4.1 reacts to MAC address change in this specific combination, Windows 2008 R2 and VMXNET3. The issue with VMware Tools causes Windows to create new adaptor in response to MAC address change, and as all newly created adaptors, it is set to DHCP.

Any other Windows version or network adaptor type combinations are not exhibiting the same behavior, and guest OS adaptor and its settings remain the same when MAC address is changed.

This was actually reported even in v4, as this affects full VM restore and replication in a similar manner.

I recommend changing adaptor type on source VM to any other. Default network adaptor for Windows 2008 R2 is still E1000 in vSphere 4.1, but you can use any other type as well, as the issue is only specific to VMXNET3. Please also note, that we had recommendation posted here from one customer suggesting to only use VMXNET3 for performance testing in lab conditions for now, as it is not quite ready for production yet (and some topics on VMware Communities actually prove that).

If you want to stick with VMXNET3, other option would be to assign static MAC address in production VM settings, this also prevents this issue as our product will preserve that as a part of backup.

[Gostev]

VMware had acknowledged the issue with VMXNET3 adaptors and Windows 2008 R2 or Windows 7:
http://kb.vmware.com/selfservice/micros ... Id=1020078

However, by this time we have already implemented a workaround (part of 5.0.1 release), so SureBackup should no longer be affected by this issue with VMXNET3 adaptors.

Thanks!

[Geniek.73]

Hi

have a question - is this workaround implemented/available also in latest version of B&R9.5? As I face the same issue with surebackup job and Windows 2008R2 VMs with vmxnet3 network interface. Or the only solution is to install mentioned MS KB or change network card to e1000 or vmxnet2 - as I was told by Veeam support? Case ID: 02416806

Dariusz

[foggy]

Yes, you need to install the mentioned MS patches to fix the issue.