Discussions related to using object storage as a backup target.
Post Reply
archvector
Enthusiast
Posts: 33
Liked: 3 times
Joined: Mar 09, 2021 7:53 am
Contact:

Clarification on connection mode for object storage

Post by archvector »

Hi,

In the documentation, I have trouble to understand the differences between "connection mode" when configuring the capacity tier of my scale-out repository:
https://helpcenter.veeam.com/docs/backu ... ml?ver=120

From my understanding :
- The direct mode is using a proxy. Are we talking here of a vmware backup proxy (the one that is installed by default on VBR server) or general backup proxy?
- The gateway mode is using a gateway. This mode is used by default and is also assigned by default to the VBR server. Is a gateway different than a proxy?

Concretely :
- My object storage is an on-premise S3 compatible storage.
- My backups and files are stored locally on my VBR server (local reFS partitions).
- I would like to push those backup to my S3 compatible storage.
Which option am i suppose to choose ?


Thanks,
Mildur
Product Manager
Posts: 9146
Liked: 2407 times
Joined: May 13, 2017 4:51 pm
Full Name: Fabian K.
Location: Switzerland
Contact:

Re: Clarification on connection mode for object storage

Post by Mildur » 1 person likes this post

Hello Archvector
- The direct mode is using a proxy. Are we talking here of a vmware backup proxy (the one that is installed by default on VBR server) or general backup proxy?
This specific statement is relevant for backup jobs. The VmWare proxy server reads the data from the vSphere environment and then transfers the data directly to the object storage.
Data flow: vSphere > Veeam Proxy > Object Storage
- The gateway mode is using a gateway. Is a gateway different than a proxy?
Yes, gateway server and proxy are two different components Gateway server means, for a backup job the proxy server will read the data from vSphere and then uses a Veeam Gateway Server to upload the data to object storage
Data flow: vSphere > Veeam Proxy > Veeam Gateway > Object Storage
This mode is used by default and is also assigned by default to the VBR server.
No, it's not the default mode. Default for new object storage repositories is "direct".
Concretely :
- My object storage is an on-premise S3 compatible storage.
- My backups and files are stored locally on my VBR server (local reFS partitions).
- I would like to push those backup to my S3 compatible storage.
Which option am i suppose to choose ?
Keep the default. It's "direct".
I assume you will use a backup copy job? A Backup Copy Job with mode "direct" will use the source repository server to connect to the object storage. Which is your VBR server:
Data flow: Source Repository (your VBR) > Object Storage

https://helpcenter.veeam.com/docs/backu ... -selection

Best,
Fabian
Product Management Analyst @ Veeam Software
archvector
Enthusiast
Posts: 33
Liked: 3 times
Joined: Mar 09, 2021 7:53 am
Contact:

Re: Clarification on connection mode for object storage

Post by archvector »

Thanks for this clarification Fabian :wink:


We have set at the scale-out repository settings the "copy backups to objects storage as soon as they are created" with a time frame configured.
Maybe that feature acts as the "backup copy job" in background
Post Reply

Who is online

Users browsing this forum: No registered users and 9 guests