Questions about S3 Endpoint - Proxmox Backup Server

[pvpaulo]

Good morning everyone.

I checked and saw that no Proxmox backup server is showing up in the S3 Endpoint, and I found this update very interesting.

Has anyone used this?

How would it work?
Does it send the backup files to an AWS bucket?
How would the restore or recovery of these backups to PVE work?

To restore, would I have to install another PBS, for example?

Or could I just take the files and restore the VMs to a PVE?

 

[Chris]

Good morning everyone.

I checked and saw that no Proxmox backup server is showing up in the S3 Endpoint, and I found this update very interesting.

Has anyone used this?

How would it work?

The S3 integration for PBS uses any S3 compatible object store to store the backup metadata and the backup data chunks it would normally store on the filesystem as backend. A local datastore cache is used nevertheless to reduce request and data volume to the s3 backend when possible (e.g. content listing).

Does it send the backup files to an AWS bucket?

Yes, but just like for regular PBS datastores, it does not store a fully self contained backup snapshot archive to the s3 object store, but rather uses a content addressable storage approach for improved data de-duplication.

How would the restore or recovery of these backups to PVE work?

By adding the PBS server with the datastore backed by S3 as a storage.

To restore, would I have to install another PBS, for example?

You can restore from the same PBS instance your are backing up to, but yes, a PBS instance is required, you cannot use the files stored on the S3 directly to restore to e.g. PVE (altough there is some advanced tooling available to recover data from the raw chunks and index files if needed).

Or could I just take the files and restore the VMs to a PVE?

No, that will not work.

 

[rwithd]

How is verification handled on S3 storage? that sounds like it could get very expensive very quickly.

 

[Neobin]

How is verification handled on S3 storage? that sounds like it could get very expensive very quickly.

Post in thread 'Surprise Backblaze S3 Costs due to egress fees and API costs'

Oct 20, 2025

Hi,

Shortly after PBS 4.0 came out, I configured a S3 backup datastore for about 600GB of data. I had verification, prune, and garbage collection jobs running daily after my backup completed. I got a surprise bill 5x my typical amount after configuring PBS to use my Backblaze S3 bucket. The verification jobs always ran for multiple hours each day, even with minimal backup change rate. As you can see in my bill, PBS downloaded 3.5TB of data over the month, even though I did zero restores. It also made 10 MILLION s3 get object requests. I also saw on my internet usage via my IPS and...

• Chris

•