Best way to configure veeam

[jbrullemans]

Hello,

I want to redisgn our veeam setup. Currently i have for all vm a separate job and we are storing the data on a NAS. I was wondering how other people are setting up veeam? I have bought an additional nas and my goal is to have two copies of every backup.

* More vm in one job
* in the job create a secondary storage space or do a backup copy job?

Also the goal of the secondary storage is that i can take it offline and put in a separate network with esx servers to do disaster recovery testing. I noticed although i'm running surebackup some backup jobs (vbk or vbm) couldn't be imported into my disaster recovery server it gave the error ""restore point corrupted".

How can i best achieve this?

[Mike Resseler]

Hi John,

There are many options that you can consider. One of the things I would consider is to create less jobs and maybe even use per VM backup files (https://helpcenter.veeam.com/docs/backu ... tml?ver=95).

Then I would indeed create a backup copy job (https://helpcenter.veeam.com/docs/backu ... tml?ver=95) to your secondary NAS.

I personally would not start with taking it offline in a separate network but continue to use our SureBackup functionality (https://helpcenter.veeam.com/docs/backu ... tml?ver=95) to do disaster recovery testing. If you indeed have issues with this as you write below, I would log a support call to understand what is going on. I personally don't like to see this "restore point corrupted" because this could mean effectively something bad... And you don't want that

Hope it helps
Mike

[jbrullemans]

Hi Mike,

For the error i already contacted veeam support unfortanly the backups couldn't be imported. What I had to do is remove old vm and compact the data. This corrupt restorepoint was because over the years we switched to a new version of vcenter. And if you looked in backups and than disk there you see the Jobs and underneath those Jobs there are multiple disks 1disk of the old vcenter and 1 disk of then new vcenter. If this is the case you cannot import it into the new Veeam 9.5. The backups are made with veeam 9. after compacting and delete old vm the backups could be imported. If i didn't do a a real disaster recovery offline i would never know this because surebackup did work.

Another question i now do reverse incremental is that still the best way?

John

[Mike Resseler]

For the reverse incremental... It kind of depends what you goal is

Reverse incremental is the best choice if the backup storage is space-constrained and if fast restore of the last backup version is priority. But it comes with a downside on more I/O and potential lower performance for the backups

There is lots of good information you can review before making a choice...

In our best practices guide: https://bp.veeam.expert/job_configurati ... thods.html
And I like this nice blog post by one of my colleagues: https://notesfrommwhite.net/2016/08/23/ ... e-written/

Hope it helps
Mike

[jbrullemans]

Hi Mike,

I have some VM about 400GB with daily change of 30 GB. I'm backing up to NAS. I think the best way is forever forward incremental. On this i get fast backup speeds and the merge is being done after the backup which cause little snapshot time instead to reverse incremental where the merge is done during the backup. As i read there i no need to do an active full. Correct?

I can do also active full every saterday but if i want to keep 14days of retention i get 3 x 400 + 21 times and 30 GB of storage space correct? Need 3 full to get a retention of 14 days.

Am i correct?

Do you know what the restore impact in time is of using Reverse Incremental vs Forever Forward incremental. Incase of disaster i want to go live ASAP after a disaster.

John

[foggy]

As i read there i no need to do an active full. Correct?

Correct.

I can do also active full every saterday but if i want to keep 14days of retention i get 3 x 400 + 21 times and 30 GB of storage space correct? Need 3 full to get a retention of 14 days.

Am i correct?

Correct.

Do you know what the restore impact in time is of using Reverse Incremental vs Forever Forward incremental. Incase of disaster i want to go live ASAP after a disaster.

On such a short chain you will hardly notice any performance impact.