VEEAM backup copy of large VM to external drive

[Rich.Philp]

Hi all,
I have a challenge that I'm not sure how to fix.

We have a 1.9TB VM that I want to backup copy to an external drive every night. This is after it does its normal reverse incremental backup.
Problem is it takes longer than I have time in the night before I have stop it cause of production disruption.

Setup details:
VEEAM backup server that is running and controlling the jobs is running on a VM.
Backup repository that also has the external drives attached (USB 3), is a physical machine.
Backup repository server and VMware server are connected via a 1GB network
VMware servers are connected to their datastore via a 10GB network.

Due to disk/nas bandwidth/traffic issues while doing backups and backup copies, I have to stop the backup copies at 8am, and start them, at 4:30 or 5pm. Backups and copies floods the VMware connection between the VMware servers and the datastores. During office hours, causes a lot of slow downs.

It works over the weekend as I can give it the entire weekend to copy.

The backup repository has the VEEAM software installed on it as well, for restore reasons mainly.
My thoughts were, but I'm not sure if it is possible: is to get the backup repository server to do the backup copy. This would make it faster, and remove it from our VMware datastore resources.

Problem I can see, is that the software on the repository server wont know anything about the backups, or even when one is happening. For it to work, it would have to import the backup before it copies.

Is there a better solution, or can I make my suggestion work?

Thank you,
Rich

[Shestakov]

Hi Rich,

Could you provide your backup job stats? It`s good to know what is your bottleneck for the job.

I would suggest to switch to forward incremental jobs with a synthetic full at the weekend.
Thank you.

[foggy]

Rich, how you backup repositories (mostly I'm interested in the backup copy job target reposotiry) are added into Veeam B&R console? Basically, when backup copy is performed, the data is transferred directly between data mover agents installed on the source and target repositories. Since both are on the same physical machine, I wonder why you're observing flood of any connection in your environment. Is it probably because repository is added as CIFS share instead of a local disk of the physical machine?

[Rich.Philp]

Thank you for your replies.

Shestakov- I will have to wait until tonight to get these. The statistics didnt give me the last run, only the current one, so no bottleneck or traffic stats.

Foggy - The Backup repositories are done using CIFS over a dedicated, 1Gb network.
Repositories are added as CIFS, not Physical drive. this is because the physical drive is not connected to the Veeam backup server, but to the backup repository machine. I hope this explains, and I understood the question.

I know if I could do the file copy from the repository machine itself, then the copy would be faster as it would be going direct to the physical drive.

Thank you,
Rich

[foggy]

Shestakov- I will have to wait until tonight to get these. The statistics didnt give me the last run, only the current one, so no bottleneck or traffic stats.

You can use left and right arrows to move between job runs and view previous job runs stats.

Foggy - The Backup repositories are done using CIFS over a dedicated, 1Gb network.
Repositories are added as CIFS, not Physical drive. this is because the physical drive is not connected to the Veeam backup server, but to the backup repository machine. I hope this explains, and I understood the question.

You need to add the physical server as Windows server into Veeam B&R console and then add both repositories as Windows-type ones (instead of CIFS). You then will be able to specify the corresponding disks/folders that store backups. This will allow for Veeam data mover agents to run locally and backup copy activity be performed locally within the repository server.

[Rich.Philp]

Thank you Foggy. I will recheck my setup and make these changes.

I might have put you wrong on the setup, but I will find out.
Is there anyway I can check to see what is doing the coping?
Thank you,
Rich

[Shestakov]

Rich,
You just left-click once on the working Backup Copy job and on the bottom of the window you can see all the actions of the job.
Thank you.

[foggy]

Is there anyway I can check to see what is doing the coping?

It's easier to check the repository type, this will tell us the data flow in your setup.