Seed a Cloud Backup for Veeam Backup Agent

Seeding is the process of creating a local backup, uploading it to the SP's infrastructure by a fast(er) internet connection (or some other method) and then mapping that backup file to the cloud backup. The primary reason to do this is the slowness of the upstream network speed of a customer, so much so that the first cloud backup (generally hundreds of GB big in size) never finishes successfully. By seeding the first backup, only the smaller daily incremental files would need to be uploaded on the slow customer network, which have a much bigger chance to finish.

Prerequisites

• The company of the customer exists in the SP console
• The Management Agent is installed on the customer's machine 

Create a normal local backup job in the SP console

Log on to the SP console (reseller or customer account) and choosing the server of which you need the cloud backup, click on Create Job 

In the Create a new job wizard, give the job a name (anything you like)

Configure it as a Server or Workstation depending the backup features you'd like to use

Choose Volume level backup if your local storage is a direct attached drive (e.g. drive E:\) or choose Entire computer if your storage is a share on a NAS device (it would look something like e.g. \\192.168.0.1\veeam) 

Under Volumes, give the volumes that you need to backup, not adding the drive that you store your backups in. Also put a tick in the Operating system checkbox.

Choose either Local storage or Shared folder depending on where your local backup storage is.

Give the destination; the retention time is not important. Click Advanced settings...

On the Storage tab, choose the Extreme compression level and the Local or LAN target (depending on your backup storage type)

Under Guest processing, choose Enable application-aware processing

Do not choose a schedule and finish the wizard.

Start the new job from the console and wait for it to run to the end.

Provide the files to your SP

At this point, you will need to provide the backup files for your SP. This can happen through FTP or any other file transferring method that you trust, using a fast upstream, or you can also give the backup files on a USB stick to one of our colleagues. ;) 

We need the whole folder

It contains two files: a .vbm and a .vbk - both is needed.

Your provider will do two things:

• download the backup files on the VCC server, into the repository of your customer
• rescan the repository so that the new backup file appears in the configuration, when you search for it in the next steps.

Mapping the backup file to a cloud backup job

Unfortunately, the SP console UI does not offer the option to map this new backup file to a backup job, so we need to do this directly through the Backup Agent installed at the customer's machine.

First, choose the server in the SP console, and choose Agent UI mode, then Switch to Full Admin Access. This is needed, because the Read-only UI does not allow adding jobs directly in the customer's Backup agent. (Read-only UI is the default for security reasons.)

Log on to the machine of the customer and choose Veeam Agent for Microsoft Windows.

Click Add New Job...

Give the job a meaningful name, like Cloud Job.

Choose the same settings as you did on the previous job: Volume level if you originally chose that, or Entire computer.

Choose the volumes and the Operating system.

This is the first point where this backup job differs from the seeding job. Here choose Veeam Cloud Connect repository

Give the DNS name of the cloud gateway.

Give the credentials of the cloud user

The following is the most important part. Click on the blue Map backup button.

The cloud repository will appear, listing the backup files that do not have a configured backup job yet. Note that the name of the file will correspond to the name of the seeding job! In case you don't see the job listed, click on the tiny refresh icon in the upper right corner.

Choose the job and click OK

Under the Advanced settings, choose the Extreme compression level and the WAN target.

No need to enable backup caching (unless you have a very strict RPO and some free space to spare). 

Enable application-aware processing

Choose a schedule for your job and configure the backup window, in case you need one, or later on, you can configure network throttling through the console, if needed and the backup window doesn't work for your use case.

Finish the wizard.

Result

Now the next time the cloud job runs, it will use the seed backup file as the first full backup, and will only create an increment. See the small restore point size in the screenshot.

On the repository side, we have the configuration file of the seed job (.vbm), the first full backup file of the seed job (.vbk) and the new increment of the cloud job (.vib) which continues the backup chain of the seed.

Post-seeding tasks

Don't forget to switch the Agent UI Mode back to Read-only UI...

...so that the backup settings cannot be modified on the customer's side.

And then delete the seed job.