Replication to Offsite Failover

[jafnva]

We have been using Veeam 11 to backup our VMs (vCenter 7) for about 2 years now. We are getting ready to stand up an offsite where we plan on replicating our VMs to. After reading about Veeam replication, where it describes configuring various failover plans and replication options, we figured that would be our best option. In the documentation, Veeam recommends using the Veeam console to bring the VMs online, but after reading on how to set this up I found that having a second Veeam backup server is not an option.

From what I understand, if our production site were to go offline then we would have to build a new Veeam server and restore the configuration backup to this server. Then hopefully be able to bring our VMs online.
Am I missing something? One of the points of having an "offsite" and replicating to it, is in case the production site is inaccessible, but I cannot find any information on how to bring the VMs online when the production site is unavailable.
Is there anyone that uses Veeam replication for this scenario?

Thanks,
John

[MarkBoothmaa]

Is there an issue with migrating your Veeam server to the offsite location?

That would probably be the simplest approach.

[tyler.jurgens]

You can have a VBR server 'ready to go' and restore the configuration to it in the event of a disaster. That way you have access to your failover plans and all backup jobs, etc.

That said, you can *simply* power on your replicated VMs. Always better to do through a failover plan since you can fail back easier, but its still possible if you power them on manually.

[jafnva]

Thanks for the replies.
Mark - the reason we want to keep the Veeam server at our Prod site is for backing up to disk. We were considering moving the backup server to the offsite but are concerned about how much bandwidth would be required and how long the backup jobs would take.

tj - that may also be an option, we were just hoping for a solution that would allow us to quickly power on the replicated VMs.

[tyler.jurgens]

What I would do then is have a secondary VBR ready to go. Have configuration backups taken regularly to a location that is always accessible by that secondary VBR.

Test an outage. Pretend the primary VBR is gone. Recovery your config file to your secondary VBR. Then test your failover plan. It really won't add much time to your DR plan.
If you want to avoid that, put your primary VBR in your secondary site and make sure you have the appropriate backup proxies setup to process your VMs from your primary site. I'd also have a secondary VBR in your primary site, just in case your secondary site goes down.

[YouGotServered]

Keep the your instance at prod site, but convert it to just a repository server, managed by your main Veeam server in your DR site. This way, you keep backups local and still have Veeam at the DR site. Just make sure you add a small repo in DR to store your replica metadata too, however.

[dimaslan]

I think what Veeam suggests in this case is to have the Veeam VBR at the production site. Each site can have their own proxies and backup repositories and the Veeam VBR's location would not matter at all.
It will simply be doing the management / orchestration of the jobs.
This way, if the production site goes down, your Veeam server is still online and you can easily execute your failover / failback plans.

[Zew]

What dimaslan is basically the recommend approach, the only thing that bugs me with that infra design is, if the offsite location experiences issues, the backup jobs themselves don't run.

[PetrM]

Hello,

Our recommendation is to keep Veeam B&R on the remote site to ensure seamless failover with minimal downtime in the event of a disaster with the production site. Please refer to the best practices guide for more information.

Thanks!