mirror backup to S3

[robyt]

Good morning, i'v some VM (~30) in four logical group.
for every group i create the backup (one full weekly and 40 incremental) and i want mirror to wasabi S3 storage.
How i can start mirroring when one of the full/incremental backup end?
I don't want to start mirror when a backup is not finished!
thank you

[acebmxer]

@robyt

Create a backup sequence.

It will ignore the time set in the initial schedules so if the previous job is still running when the next schedule job is set to start. This will hold the next job until the first job completes.

[robyt]

@acebmxer ok, i don't use (for full + delta) the old schedulng (one job but two scheduling, one full and one delta), but i must separate the two jobs?

[acebmxer]

@robyt

Unless im am not truly understanding what your asking.

I have 1 backup job. That job has 2 schedules. One for daily incremental backups and one that has the force full backup checked. The full backup only runs on Sunday. The incremental does not run on Sunday.

When creating the sequence job it uses the schedules from that job to create the sequence job.

[DustinB]

@acebmxer Not a backup job, a backup sequence.

Jobs are for creating separate backups on (usually different schedules).

Sequences are Backup and then do XYZ.

[acebmxer]

@DustinB

Yes but you cant create the sequence with out the schedules from a regular backup.

[robyt]

@acebmxer [excuse for the poor english!]
i've now this situation:
1 backup job with two disables schedules, one full and one delta, to a nas
1 mirror full backup to wasabi (S3)
1 mirror incremental backup

i've insert two sequences:
one starting at sunday for full backup (the sequence is full backup and then full mirror)
one every 3 hours with delta backup and then mirror incremental

the job start at the correct hour but the mirror incremental send every time the same data size..
backup to nas:

dns_interno1 (ctx1.tosnet.it)
Transfer data using NBD
    Clean VM directory
    cleanVm: incorrect backup size in metadata
    Start: 2025-06-24 16:00
    End: 2025-06-24 16:00
    Snapshot
    Start: 2025-06-24 16:00
    End: 2025-06-24 16:00
    Backup XEN OLD
        transfer
        Start: 2025-06-24 16:00
        End: 2025-06-24 16:01
        Duration: a few seconds
        Size: 132 MiB
        Speed: 11.86 MiB/s
    Start: 2025-06-24 16:00
    End: 2025-06-24 16:01
    Duration: a minute

Start: 2025-06-24 16:00
End: 2025-06-24 16:01
Duration: a minute
Type: delta

 dns_interno1 (ctx1.tosnet.it)
    Wasabi
        transfer
        Start: 2025-06-24 16:02
        End: 2025-06-24 16:15
        Duration: 13 minutes
        Size: 25.03 GiB
        Speed: 34.14 MiB/s
        transfer
        Start: 2025-06-24 16:15
        End: 2025-06-24 16:15
        Duration: a few seconds
        Size: 394 MiB
        Speed: 22.49 MiB/s
    Start: 2025-06-24 16:02
    End: 2025-06-24 16:17
    Duration: 15 minutes
    Wasabi
    Start: 2025-06-24 16:15
    End: 2025-06-24 16:17
    Duration: 2 minutes

Start: 2025-06-24 16:02
End: 2025-06-24 16:17
Duration: 15 minutes

the job send every time 25gb to wasabi, not the incremental data.

[acebmxer]

@robyt

What i posted was just examples of how i have my backups configured. You dont have to call the backup jobs the same as mine. Maybe someone else with more experience can step in. Double check your schedules make sure you did set incorrect settings in one of the schedules/jobs.

[robyt]

@acebmxer of course, this is only a test.
the problem is not the schedulng but why incremental send every time all data.

[florent]

@robyt said in mirror backup to S3:

1 backup job with two disables schedules, one full and one delta, to a nas

Hi robyt

i the mirror settings : full mirror if when mirroring full backup ( called backup ) , as one file per VM
The incremental mirror will take care on all the files created by a delta backup

Can you show me what retention you have on both side ? you have more explanation on the syncrohnization algorithm here https://docs.xen-orchestra.com/mirror_backup , to be short a delta mirror can only be done if part of the disks chains are common to both remotes

[robyt]

@florent hi, i've adjusted the retention parameters and i'm waiting for some days of backup/mirror for checking