Mixing VEEAM AWS intergration features

[Almero]

Good afternoon communities .

I am having difficulty defining and end to end solution using two separate VEEAM + AWS features .
( Azure would also work , but AWS looks more streamlined for direct restore as there is no conversion of vmdks )

We run vSphere 6.5 on prem and use VEEAM9.5u4 with about 710VMs , mostly Centos , with some Windows .
Roughly 130TB used for all tiers , PROD , DEV and UAT with limited block based backup repo space.

I have two requirements going forward , but they appear to be two separate VEEAM features .

1 > I want to use SOBR Scale out to AWS S3 IA for longer retention and off site backups.
This appears easy enough to do , although RPO of closed off backup cycles would be more than a week old under default behavior .

2> I want to use AWS EC2 as a DR site for on prem vSphere .
From KBs it appears EC2 restores cannot be done from S3 object based , and requires me to push from on prem repos . ( On prem that could be lost/down during DR invocation )

Options appear to exist where I can spin up an expensive $ per GB EC2 EBS VM , and use that as a secondary target for daily backups .
But that would cost a lot more than just S3 , and could even mean i need to push data upstream x2.
Once to S3 for all VMs to have longer retention using SOBR .
Once to EBS/EC2 for my DR critical PROD VMs using secondary target on top of my on prem copy during backup job .

Is there a way to "mount" the S3 object based storage from EC2 in the event of DR ?
This is certainly possible from an AWS point of view , but from what i can gather VEEAM cannot restore directly from that source .
Also i believe a lot of the metadata would still sit on my on prem repo .

In other words , can i use Scale out capacity repository , and use that same copy of data as continuity restore option ?
Fortunately enough our apps are fairly stateless , and continuity SLA is 7 days . ( RPO = 7 Days , RTO = 2-3 Days )

Can anybody guide me towards an end to end hybrid model in current versions of VEEAM?
It appears v.10 could answer some of these questions .

Thank you in advance .

[veremin]

2> I want to use AWS EC2 as a DR site for on prem vSphere . From KBs it appears EC2 restores cannot be done from S3 object based , and requires me to push from on prem repos . ( On prem that could be lost/down during DR invocation )

Can you provide the KB article mentioned? I'm wondering cause it's absolutely possible to restore VMs to AWS EC2 from backups stored in Capacity Tier. Thanks!

[Almero]

Hello , if that is the case , that is wonderful . The wording in this KB led me to believe it must be from my on prem . Reading it now , it does not clearly state that .
However , is metadata not mostly stored in my in prem device ? So in the event i lose my on prem DC ( FULL DR ) , this might not work

https://helpcenter.veeam.com/docs/backu ... l?ver=95u4

[veremin]

Correct, you can restore VMs to EC2 from backups stored in Capacity Tier - no issues with that.

If you loose your on-prem environment, you will first need to:

- either recreate backup server by restoring its configuration
- or deploy new backup server and recreate Scale-Out Backup Repository along with Capacity Tier

After that, you will be able to restore VMs to EC2.

Thanks!