Backup of NAS, file shares, file servers and object storage.
Post Reply
mike7891
Influencer
Posts: 17
Liked: 1 time
Joined: Feb 15, 2025 9:15 pm
Contact:

Troubles with S3 repo Mount Server and Backup Copy jobs

Post by mike7891 »

Hello, I am running into a bit of a roadblock regarding an S3 repo Mount Server and how VBR chooses its connection for Backup Copy Jobs from the S3 repository -> on premise Hardened Linux Repository. Let me explain what is going on.

1. We have two main environments, one in AWS and another on premise. A site-to-site VPN exists between these sites. Our VBR server is in our on premise location.
2. Currently we use Agents deployed via Protection Group to the servers up in AWS EC2. They back up to an S3 repository using the Direct connection mode (no gateway server selected). The site-to-site VPN is established for management traffic so that the VBR server can deploy the agents to the EC2 servers.
3. A Backup Copy Job exists that copies these backups out of S3 to our on-premise Hardened Linux Repository. This backup traffic does not go over our VPN, but direct between the VBR server and S3 as our dedicated WAN connection is much faster than the VPN.

So far the above works great! We get EC2 backups in S3 and nightly copies down to our on premise Hardened Linux repo. The problems arise once I have tried to add a Mount Server for use with FLR and Application aware restores for the S3 repository. The purpose of this was to deploy a Mount Server up in AWS EC2 so that we could do FLR from the backups in S3 to EC2. However now that the Mount Server exists, the Backup Copy Job insists on using that instead of the direct WAN connection, causing the copies from S3 -> Hardened Repo to go over the VPN. Is there any possible way to coexist the Mount Server while still having the Backup Copy job use the direct WAN connection as it does when the Mount Server is not present? It seems no matter what I've tried, the Backup Copy Job is forcing itself to use the Mount Server.

I understand this use case is uncommon. Most people copy from on-premise to S3. We are doing the opposite, so maybe this type of situation is not really supported how I am envisioning. I appreciate any insight on this!
Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest