pve-zsync: reduce number of snapshots on source, while keeping all on destination side?

[mailinglists]

Hi,

I have a backup system set up using pve-zsync. It works perfectly.

The only issue I have is that the number of snapshots created on source is the same as on destination.
So If i want to have 12 monthly backups i have the data for all 12 monthly backups on source as well as on target server. Obviously i would rather to free up space on the source server and keep the older copies only on backup server. Let me know, if I was not clear enough with my wishes.

So the question is, is there a way or a possible workaround to have diff backups (with ZFS), that would leave only the last few snapshots on the source, while keeping all older on the destination / backup?

 

[WhiteStarEOF]

I've not used pve-zsync myself, so I can't answer for it.

However, I do use Sanoid to sync snapshots between Proxmox boxes, and even Proxmox to FreeNAS. Maybe it will be able to help you as well. The configuration for snapshot retention is set on each box, so you can have the source keep only say 2 weeks worth of snapshots, and the destination keeps 6 months.

https://github.com/jimsalterjrs/sanoid/

 

[udo]

Hi,

I have a backup system set up using pve-zsync. It works perfectly.

The only issue I have is that the number of snapshots created on source is the same as on destination.
So If i want to have 12 monthly backups i have the data for all 12 monthly backups on source as well as on target server. Obviously i would rather to free up space on the source server and keep the older copies only on backup server. Let me know, if I was not clear enough with my wishes.

So the question is, is there a way or a possible workaround to have diff backups (with ZFS), that would leave only the last few snapshots on the source, while keeping all older on the destination / backup?

Hi,
unfortunalty this don't work with pve-zsync.

But you can use an different tool: znapsend. With this you can create an migration plan - like every hour sync, keep one snapshot each day (since yesterday), and keep one snapshot for each week after that... and this different on the target and source node.

Very flexible, but must be configured for each vm-disk...

I use it for disaster recovery. Work well most of the time.

Udo

 

[mailinglists]

Thank you guys. Looks like I will migrate to one of the before mentioned scripts.