High Level Disaster Recovery and testing

[andyboy16]

I am trying to come up with a high-level DR plan from on prem to AWS. Our current environment consist of the following:

-Datacenter running VMware with 100+ VM’s running a mixture of SQL and Oracle. We only backup database servers for the most part with application awareness enabled.
-Veeam B&R is running inside our VMware environment and backups are stored locally on SAN and copied out to S3 using the SOBR method. Copy of backup config file is stored on S3 every night.

Our backup plan is to restore databases to AWS RDS instance (app servers are easily restored with just config files)

Question #1:
If the entire datacenter blew up and collapsed and everything was gone, what would be the best method in restoring? Would standing up a Veeam EC2 instance, importing backup configuration file and then restore DB items to RDS instance work? Even if our entire VMware instance is gone would this be the correct method? Would restoring from Veeam to RDS even work? I’ve tried searching restore to RDS but can’t find any info about this topic.

Question #2:
If we wanted to test this scenario, could we currently stand up a Veeam instance in EC2, import the backup config file, and test this out? Will importing the Veeam backup config mess up our “production” Veeam backups/configuration?

[HannesK]

Hello,
Q1: "Veeam in EC2 -> import data -> start restore" would be the easiest way that comes into my mind. Because configuration restore does not really sound useful to me, because none of the components exists in AWS. And VMware jobs don't work in AWS etc. There is no "restore to RDS". So "restore to EC2" would be the only option. I strongly recommend to make yourself familiar with that type of restore (it takes time...)

Q2: you can restore the configuration. Network connections to on-prem fail anyway (except you have a network connection) The only important thing: only import the object storage. Don't create a scale out repository (SOBR). Because two backup servers accessing the same SOBR object storage bucket cause issues.

Best regards,
Hannes

[andyboy16]

What if we stand up a EC2 instance in AWS, import the object storage and then restore SQL via "Restore application items" for SQL Server database to a RDS instance. Would that not work?

[HannesK]

There is no "restore to RDS"

the Veeam Explorer for SQL has no option to restore to RDS. One has to do that with native AWS methods.