B&R Server at DR site with Copy Jobs only

[JoshuaPostSAMC]

I have a good handle on Veeam on-premise backing up to a local respository. I'm preparing to start Backup Copy Jobs to store backups in a DR site as a second respository. At this time we are not going to use full replication for all VMs, although that may come in a future phase.

Can someone explain or point to me to an article about what I should do for a backup server at the DR site? In the event the primary datacenter is completely down, we will have all of the repository data at the DR site and the network IPs will all be the same, but what should I do for B&R Servers for starting to recover?

My original plan was to have a second B&R server at the DR site that has the same folder setup as the repository, and I would need to complete a rescan of the repository before starting restores. Hopefully this could be scripted to happen each morning as well. I saw one forum post about just replicating the B&R server, which sounds like a better idea, so I would just power on the replicate at the DR site and as long as replication happened after the last job completed, then I shouldn't need to rescan the DR repository and can just start restoring to the vCenter at the DR site.

Is that the best plan, or is there another one I haven't thought of?

[fraack]

Hi Joshua,

you have several options. Why are not run your backup server at your DR site?

If your VBR server is a virtual machine you could replicate it to DR site. If you have problems in your primary Datacenter you could power on the replicated VBR manuelly.

Many of my customers are working with physical VBR servers. In most cases we have a physical machine in the DR site as well. Than we place our configuration backup at this machine and keep the latest Veeam installer files on it. In case of an emergency we are able to install Veeam Software and restore the config backup file. That works very well.

[JoshuaPostSAMC]

Our backup server is virtual, saving to a CIFS target for the primary repository, so running it primarily at the DR site would slow down our standard backups. Since my understanding is that the copy jobs read from the repository, VMs won't be running on snapshots as long with this model.

I am liking the idea of doing replication for the B&R server. I have the SQL database installed locally so it can be portable, and this looks to be my best option