Hi guys,

I got an XCP-ng Server that I I do backups for like 11 vm.
I use XO from source.
I’ve a Synology has a nfs remote.
And I’m using backblaze for my backup. Mirror.
On the nfs, I use multiple block feature.
On my s3 this features is not there so I don’t use it.
On my synolgoy, ive 2.5tb of data.
On my Backblaze, I’ve 11tb.
I don’t know why.
I saw that the structure of folders is:
Xo-vm-backup -> UUID-VM -> vdis -> UUID-BACKUPJOB -> UUID-disks -> data -> UUID-INCREMENTAL -> blocks (something like that I think)
I do see that I’ve a lot more folders on Backblaze in the data folders (the UUID-INCREMTAL folders I think.)
I’ve 169 in Backblaze for 1 vm and 45 on the synology.
I did some modifications on my jobs since that because it didn’t run has I wanted and removed some vm because of that because I didn’t want the to be backup on Backblaze.

I just changed the option “Delete Before” on the mirror job.
Since my objective is to not over kill my remote data, its not that much of a deal to lose 1 or 2 increment if my backup doesn’t work once or twice.

My questions are:
Will it delete those over incremental that’s not supposed to be there if I select the delete first?
Should I use the “Store backup as multiple data blocks instead of a whole VHD file. (creates 500-1000 files per backed up GB but allows faster merge)” if i use a mirror backup to backblaze (a s3 compatible storage)?

Do people had issue where the backup mirror is far bigger than the source backup?

I see an older job in backblaze for 1 vm. In health check, its not there. Will it delete that one too?