Large Backup copy Job - Overlap primary Backup

[blublub]

Hi

I have set up backup copy jobs to move backups to our OffSite NAS. This works fine and timely for the smaller VMs.

However we have one larger VM of about 35TB. The copy job will need about 15 days over the 400mbit connection.

backup copy Job:
Copy VM 01 to Offsite -> direct mode -> prune mode, copy last restore point only

What will happen if the daily primary backup job of that VMs runs/Starts?
Backup VM 01 - runs daily at 01:00
Backup mode: reverse incremental
Keep restore points: 3

So what will happen every night at 01:00 when the primary job runs? - from the help section I guess the backup copy will be paused when primary backup starts nad is running - correct?

What will happen when the primary backup has finished and hour or two later? - will it restart completely as the the last backup file has changed, (reverse incremental), it might never finish?

Will it know what has changed the last two hours and account for that? - after 14 days, will it have the copied the restore point from 13 days ago when I started the copy or the "current" one due to "awareness" what has changed? (3 restore points in primary are only 3 days, so primary wont keep data for 13 days...)

thx

[david.domask]

Hi blublub,

It will work as the User Guide mentions. Since you're using Reverse Incremental, I did run a few quick tests just to be sure as Reverse Incremental merges data in and out of the Full Backup at the start of the chain, but there was no issue, the Backup Copy waited as expected.

However, this might be a situation where Backup Copy Seeding is best, as it's designed to help avoid the situation you're discussing where the initial transfer might be a long time even with a decent connection. I would advise consider seeding as opposed to waiting it out; there's no issue with either, just I think seeding will be the better option for you.

[blublub]

OK, thx - so every day when he primary backup job is running the copy jobs pauses and continues, thats good.

What does the backup copy job do with the new data? - since its a reverse incremental I would assume that the copy jobs tries to stay in sync and will add chaged blocks every day?

A seed would require extra, portable drives which I do not have at hand since why we are paying for a 400/400 connection at both sides

[blublub]

Ok, here is what happens:

Primary Job - reverse incremental, 3 retention point, runs daily - volume 32TB

Backup copy job over LAN/WAN will take more than 24hrs - every day when the primary job runs the backup copy pauses and resume. However after 3 days there is a problem since the backup job started on may 30th and the primary job has only 3 retention points, so may 30th will be deleted by the primary job and the copy job will fail as its task was to copy the data set from may 30th

So with large VMs this can lead to issues during the first backup copy process, either adapt the retention policy for the primary backup job or pause it until the copy job has finished