Installing second instance of B&R at DR site

[leetansey]

I am currently in the process of creating a backup copy job to duplicate our onsite backups to an offsite location, which will also serve as a DR site. Following the initial seed and with new restore points being copied as they appear, I need the ability to perform a full recovery of all VMs contained on the offsite backup server, should the need arise.

Am I able to (from a licensing perspective) install another instance of my licensed Veeam Backup Essentials software onto the offsite server containing the backup copies, or would I need to install Veeam B&R Community Edition instead?

Many thanks,

[Mildur]

For restore purposes, Community edition should do it

It should be installed in a standalone vm/server.
You can‘t install a new Veeam VBR server on top of a existing Veeam Backup Repo Server of another Veeam VBR Environment. In emergency cases, it‘s ok.

Best Practice in your scenario should be to install your main Veeam VBR in the off site datacenter. And manage all your backups from there.

Main Backup Repo server can be in your production site.
Backup Proxys can be in your production sites.

Second backup Repo is in your DR site (offsite).
Veeam Management Server (VBR) is in your DR site.

[leetansey]

Thanks for the quick reply.

Just so that I understand and to explain the scenario further, I have provisioned a DR server in a branch office running Windows Server 2019 that holds the backup copy data for the main office VMs on a shared folder, as the two locations are connected through a site-to-site VPN. The DR server currently only has the essential Veeam software components installed that allow it to be part of the backup infrastructure and to act as a repository.

With the DR server now in this configuration, is it not possible or advisable to install Veeam B&R Community Edition to facilitate the recovery of the VMs onto the same server once the Hyper-V role has been installed?

Many thanks,

[Mildur]

Is this shared Folder an SMB Backup Repo?
Then you will have no veeam components installed on It. Veeam do not install software on the smb target. Only on a gateway server, if you have configured one. If not, your Veeam B&R will be used.
If thats the case, Installation of Veeam B&R Community Edition should be no problem.

If you install Veeam V&R Community Edition on a server, which has already installed software components from another veeam server, then the Backup repo configuration on the DR Server can be overwritten by this new installation. The production Veeam B&R can loose access. This is what I think will happen.
Never share a single windows server with two different Veeam B&R at the same time.

————

Another thing I would do:

If you have an DR Site, look into Veeam Replication feature.
It‘s better for short RTO if a disaster is happening.
You will have your vms ready to boot up, if production site is not running.

[foggy]

If you install Veeam V&R Community Edition on a server, which has already installed software components from another veeam server, then the Backup repo configuration on the DR Server can be overwritten by this new installation. The production Veeam B&R can loose access. This is what I think will happen.
Never share a single windows server with two different Veeam B&R at the same time.

Veeam B&R components can be shared between different installations provided they are at the same patch/build level. Though this is not generally recommended for various reasons.

[Mildur]

Hi Foggy
Thanks. I wasn‘t aware of that
I will stick to your reccomandation „never share veeam b&r components“

[leetansey]

Thank you both for your feedback. I think I will either install VBR on a separate machine at the DR site that has access to the SMB backup repo, or look into replication for reduced RTO instead.

[Mildur]

No problem Happy to help.
Have a nice weekend and stay healthy

[foggy]

I will stick to your reccomandation „never share veeam b&r components“

This is mainly due to the load considerations since different Veeam B&R instances are not aware of each other so will assign tasks to the same server without taking into account the second instance's concurrent tasks limit.

[leetansey]

I have since looked at implementing the replication feature for creating replica VMs at the DR site, however the VBR server is located at the primary site along with the primary repo. If the primary site was wiped out then I would not have the VBR server available to perform failover of the VMs, network and IP remapping. Can anyone advise on whether it is possible to achieve failover and failback in this scenario? Since different VBR instances are not aware of each other, any instance of VBR at the DR site may not be of benefit.

Many thanks,

[Vitaliy S.]

To perform network and IP remapping, your Veeam B&R server has to be available. In case of the source site destruction, the only available option would be to use the secondary Veeam B&R server (that controls job replicas) or Veeam Availability Orchestrator to perform the failover with all required steps. Thanks!

[leetansey]

I don't have a secondary B&R server at the DR site at present as I want to get the general approach right before going ahead. If I implement replication to the DR site onto a server running the Hyper-V role and the primary site is destroyed, am I correct in thinking that the only option would be to manually start the servers through Hyper-V Manager and perform manual network and IP remapping? Is there any benefit to having a second B&R instance at the DR site apart from backing up the VMs again when moving back to the primary site when it is available again? I was under the impression that the backup repository, hosts and backup servers for replication all had to exist at the source site, and VBR installed at the DR site would therefore not be able to control the replication job?

Many thanks!

[Vitaliy S.]

If I implement replication to the DR site onto a server running the Hyper-V role and the primary site is destroyed, am I correct in thinking that the only option would be to manually start the servers through Hyper-V Manager and perform manual network and IP remapping?

Yes, that's correct.

Is there any benefit to having a second B&R instance at the DR site apart from backing up the VMs again when moving back to the primary site when it is available again?

Yes, the biggest benefit is the ability to perform a controlled failover with re-IP etc.

I was under the impression that the backup repository, hosts and backup servers for replication all had to exist at the source site, and VBR installed at the DR site would therefore not be able to control the replication job?

You need only proxy servers at the source site in the replication scenario.

[foggy]

You can install a second Veeam B&R instance at the remote site, create replication jobs there and do a so-called "pull" replication (unlike "push" replication which is done by the source-side instance). This will allow you to perform failover from Veeam B&R console instead of starting the VMs and doing re-IP manually. The source Veeam B&R can still be responsible for running the backup jobs. This is a pretty common approach among our customers.

[AllenE]

Hey Guys, with the same main site failure scenario and have a replication setup (with re-ip) to an esxi host at DR. Will I be able to perform a controlled failover with re-IP if I spin up a new VBR at DR and restore from a configuration database backup?. Currently, this is roughly the setup and the recovery scenario. I just feel that this is not very efficient and I'm thinking of changing the main VBR placement to DR site or perhaps use it to manage a "pull" replication job.

[foggy]

Yes, you will be able to do that.