VEEAM 9 DR Scenario

[RhysHead]

Hi,

I have a bit of a newbie question surrounding VEEAM backup & replication and a DR scenario that I cannot seem to find the answer too.

Basic setup is as follows.

VEEAM 9 used to replicate production to an offsite facility with re-ip setup for the DR site.

I understand than for the re-ip to take effect, the failover needs to be initiated from the VEEAM server.

The scenario I have is that the production site becomes unavailable completely (Catastrophic failure).

How would a failover be initiated when the VEEAM server is no longer available?
Is there a way of automating a failover in the event of the production site going offline?

Thanks for any help

Rhys

[foggy]

Hi Rhys, the recommended approach to replication is to have Veeam B&R instance responsible for replication jobs residing in the target location - in that case you will be able to perform automatic failover just fine. Another way of addressing this is backing up Veeam B&R configuration, importing it into a newly set up Veeam B&R instance in case of DR, and performing failover using this new instance.

[RhysHead]

Hi Foggy,

That is what I suspected.

If the VEEAM was moved into DR and failover controlled from there, when the failback was initiated would it ensure that any data changed in DR will then be replicated back into production?

for a test, the process is going to be manual, whereby the link will be cut to DR, the DR machines powered up and the IP's changed manually, however I suspect that when the production is brought back online then it will replicate back to DR, thus
destroying any changes that have been made during the DR test.

Is there a way to prevent this from happening therefore allowing the changed data in DR to be replicated back to production?

Thanks
Rhys

[foggy]

If the VEEAM was moved into DR and failover controlled from there, when the failback was initiated would it ensure that any data changed in DR will then be replicated back into production?

During failback, changes occurred in replica VMs while they were running are synced to production. This is what you want to do to avoid this data loss.

for a test, the process is going to be manual, whereby the link will be cut to DR, the DR machines powered up and the IP's changed manually, however I suspect that when the production is brought back online then it will replicate back to DR, thus destroying any changes that have been made during the DR test.

But do you need these changes occurred in replica VMs residing in DR, if all this time your production VMs were up and running and used normally (if we're talking about DR testing)? You just need to discard the DR changes, which occurs automatically during the next replication jobs cycle.

Is there a way to prevent this from happening therefore allowing the changed data in DR to be replicated back to production?

Not sure why you need this in case of DR test.