Sharepoint restore environment

[Fizzer]

Hi All,
My first post here, we've just purchased Veeam, so please be nice!
I have set up a basic environment to start feeling my way into the product.
So far I have successfully backed up, destroyed and recovered an Oracle Windows VM using VSS - very impressed with Veeam.
Later on I will be looking at expanding the Veeam environment to improve performance.

However, my first and most important task is to backup and restore items from our Sharepoint 2010 farm. I have a test farm which I have backed up. It consists of:

1 SQL server
2 Back end (search) servers
2 front end servers
1 DC

The back end and front end pairs are network load balanced.

My first assumption is that I need to boot all 6 of these servers into a Virtual Lab environment.
I've tried using surebackup to do this, however this seems to fail when pinging, fail to detect dhcp or DNS services. The timeout values seem to be quite long. This leads me to believe that it might take anything up to an hour to boot all 6 servers before we can start restoring anything when I know they should be able to boot in minutes.

Would it be possible to create an environment similar to the Surebackup Virtual Lab where it just creates an isolated environment then boots all 6 VM's simultaniously without performaning all the checks? This would enable me to interact with the environment via a console and extract restored sharepoint items/lists sites etc.

Thanks in Advance.

Andy

[foggy]

Andy, make sure you have an application group with your DC, SQL, and DNS servers (actually, all VMs on which VMs to be fired up from backup are dependent) configured. In application group you can specify roles for every VM, boot priority and boot delays, so all services required for restored VM operation are started in the correct order.

After starting VMs from the application group, Veeam B&R will then boot VMs from the specified backup and you will be able to restore items.

Should more information on SureBackup workflow required, please have a look at our sticky F.A.Q. and product documentation.

Thanks.

[Cokovic]

And as always maybe it's a problem with Windows Server 2008 R2 and VMXNET3 NICs. So please also take a look at this KB article from VMWare.

[Fizzer]

Hi Folks, thanks for the info.
The DC is 2k3 so no problem with vmnic (thnx for KB link though, will keep that in mind when we upgrade AD)!
I have set up an application group for sharepoint with the DC, SQL and 4 sharepoint servers in.
I have applied roles for DC, DNS and SQL to the appropriate servers
I haven't changed the timeout values yet.

The problem seems to be that the DC (first in the list) fails the ping test rule. If I try without ping test, it gets through to 'Running Test Scripts' successfully then fails Domain controller AND DNS server tests with 'error code 10060'

I wonder if the helper VM is unable to communicate with the VM in the isolated network, preventing pings and testing of services.

As an aside, is it possible to boot all VM's in an application group simultaniously?

[Cokovic]

Maybe then its a firewall problem. Or the setup of your lab is wrong. Can you ping the proxy appliance from the Veeam server where the SureBackup job is started from?

As an aside, is it possible to boot all VM's in an application group simultaniously?

As far as i remember no. You could use a linked job to run some VMs simultaneously. But the VMs in the application group still got started first in a sequential order. And you cant keep the VMs from the linked job up and running (after the test they got shut down immediately).

[Fizzer]

yeah, maybe there's a networking problem I need to look at - (a job for next week I think)

It seems a shame that I can't fire up several VMs in a lab all at once. For our 6 server sharepoint environment it's going to take a long time to bring the farm up to restore the smallest file but just a few minutes to recover the entire farm in a DR situation.

I'll keep you posted on progress, thnx for the help so far.

[Cokovic]

Opening a support case is a good idea too as they can assist you in reviewing your logs and taking a look at your setup over webex

[Vitaliy S.]

Hi Andy,

Actually you do have an ability to spin up several VMs in a lab at once, please check out this thread for more info: Add vms to a running virtual lab?

Hope this helps!

[Cokovic]

Uh yeah forgt about this method as it never came to my mind doing it that way

[Fizzer]

genius.
thank you

[Fizzer]

Just having a look around for Veeam Powershell documentation, does it exist?

If not, could anyone help me with a script to Instant Recover 6 VM's onto an isolated virtual switch and power them on (network connected)
I guess it would need to prompt for the backup point too.

[foggy]

Andy, you can get the PowerShell reference guide at the Resources tab of the product web page.

In case assistance is required, please use the corresponding PowerShell subforum to post a request, and other community members may be able to help you with this.

Thanks!

[Fizzer]

thanks Foggy, found it.
Also posted on the powershell forum as it's not my strong point.