Virtual Lab assistance and Questions

[ricenhower]

Newbie to Veeam...

Product: Veeam Backup and Replication v8

My company would like to create a test environment that mimics a handful of production servers. This would be a tear down and rebuild upon demand that may stay up and running for months and on occasion it could becoming a staging environment for major upgrades or revisions.

I was originally going to use SureBackup and On-Demand Sandbox to accomplish this leveraging Veeam’s technology to isolate the network using the Virtual Appliance to provide IP masquerading but discovered that this is a “READ ONLY” environment running from Veeam’s backup repository and writing changes datastore location you define during setup. All changes could be lost if the Virtual Lab is powered off and I’m also concerned about degrading the performance of my Veeam backup repository.

Please correct me if I’m wrong in my understanding of the technology above.

Does SureReplica also run as a “Read Only” environment?
Is there a way to create this environment so we can read/write to the replica .vmdk opposed to Veeam backup file?

[PTide]

Hi,

Please correct me if I’m wrong in my understanding of the technology above.

That's correct, both jobs do not allow to commit changes in order to prevent backup/replica corruption and produce delta-files that grow during SureBackup/Replica jobs. If you want to have some kind of a staging lab then you might want to try the following workaround:

1. Set up a Virtual Lab with a proxy appliance
2. Clone your production machines and attach them to Virtual Lab networks with "internal" IPs assigned.
3. Enjoy your new lab, running from real .vmdks, without growing deltas =)

The only drawback is that you need to configure routing by yourself (in case you want your staging environment to interact with production). That can be accomplished if you login into proxy appliance and set up your own rules.

[ricenhower]

Thanks for your reply...

Would there be any benefit from using the Veeam replication job to create the environment pulling from the last backup opposed to cloning?

I've never gone through this process so, could you expand on what you mean about attaching them to virtual lab with internal IP's assigned.

[PTide]

Would there be any benefit from using the Veeam replication job to create the environment pulling from the last backup opposed to cloning?

All you need is just to recreate your production VMs inside an isolated network, no matter how.

I've never gone through this process so, could you expand on what you mean about attaching them to virtual lab with internal IP's assigned.

On a stage of configuring a virtual lab there is an option to setup isolated networks (in case you choose "advanced single host"):


There you can create isolated networks inside your lab and map them to production networks. By that you basically add some production-alike networks into your virtualization environment and connect them to your proxy appliance therefore letting proxy appliance server as a proxy between lab and production:




Next, you can have masquerade IPs configured in order to let your production communicate with your lab, if needed:



In the end, you need to place your cloned/restored/replicated/whatever VMs into isolated networks you've created during previous steps. If you run your SureBackup Job you can see that it's done automatically (see that VM with a long index?):


Also please see the corresponding section of helpcenter for more detailed explanation of how virtual lab works.

Thank you.

[ricenhower]

Very much appreciated

[kinto]

Hello,
I've a question about the Virtual Lab: is there any configuration I can use to have a permanent Virtual Lab (until I want to delete it).

I'm trying Veeam B&R on a demo installation (Veeam server and Hyper-V server on the same machine) and after a power failure I've lost a V-Lab I was using.
What I need is the possibility to start a Sure Backup Job and to restore it also after a reboot or shutdown of this machine.

Is that possible?

Thank you.

[PTide]

Hi,

What I need is the possibility to start a Sure Backup Job and to restore it also after a reboot or shutdown of this machine.

Yes, some kind of a permanent Lab is possible. Please see this message.

Thank you.

[NPOrange]

Hi,

Would it be possible to tell Veeam SureBackup/SureReplica job to fully restore a VM to another Datastore instead of using Instant Recovery technology ?
Sure, the restore phase would last longer and would use a lot of space but then it would make it possible to test backup over a dedup appliance (StoreOnce, DD, ...).
Furthermore, if the vLab run for multiple days, maybe it'd have a lesser impact than the use of vPowerNFS/Instant Recovery ?

Is it something that is foreseen for next versions ?

Best Regards,

Nicolas

[PTide]

Hi,

It seems that you might be interested in a permanent vLab, please check this post.

Thanks.

[NPOrange]

Thanks for the very detailed answer. Yet, it would be great to have it automated with a simple checkbox at the job's creation !

[Fwiler]

I finally created a virtual lab that works. I wanted to make this permanent though.
I found I couldn't replicate or backup the virtual lab vm's with Veeam.
Could I export the vm's including the Virtual Lab vm, stop the SureBackup, then import the vm's?

[foggy]

Dustin, please see above for instructions on achieving what you're after.

[DavidReimers]

I'm in the process of setting up an SRM-like equivalent for Veeam replicas.

We don't have a single ESX host with sufficient capacity to run all the VMs we want to test.
Do we need a dvSwitch to achieve a multi-host configuration, or would we be able to leverage our existing 'sandpit' VLAN and vSwitches?

[PTide]

Hi,

You can schedule several Surebackup jobs to test the VMs by groups. If the VMs constitute some milti-tier application and you want to run them simultaneously then you need dvSwitch. Also, would you clarify what you mean by SRM please?

Thanks

[DavidReimers]

We don't have ESX Enterprise Plus licensing, so dvSwitches are a no-go.

SRM = VMware Site Recovery Manager.

We are going to follow this procedure: http://www.virtualizationteam.com/bcdr/ ... edure.html
Our DR site has an existing sandbox VLAN, so it we will configure our ESX vSwitches for an additional network, then connect the replica VMs to this network, so they can talk to VMs on other hosts.