From the many challenging tasks IT departments have to navigate when managing the modern data centre, data availability and data recovery are some of the most critical. The reason is what Gartner calls the “Data Explosion”. As humans living in the digitised world,
we generate enormous quantities of data, every second of every minute, every day. This rate of data generation is a challenge for the IT department in so many ways; it all needs managing, storing, and backing up.
The 3-2-1 data backup principle is a highly recommended backup design. The three is the three copies of your data, the two is two copies on different media, and the one is the backup off-site; but the challenge of the one is how to do the off-site backup. How do you initially move what is likely to be a significant amount of data off-site through your likely limited bandwidth? A good start is to begin with “Data Seeding”, or setting up with cold data to speed the uptake of your production data on your off-site backup target.
I have received many requests from customers and Service Providers for an easy to understand explanation to setting up data seeding. In this article, I will take you through the process of seeding Veeam backups to off-site repositories, especially if you are using Veeam Cloud Connect.
Pre-Requests
To follow the steps in this article, you must be familiar the following products, and their configurations:
- Veeam Backup and Replication,
- Off-site Repository (Connection to a Service provider),
- External Hard Drive or Tape Drive/Media.
The Challenge
As listed above, you will by now have your Veeam Backup and Replication product configured and will be signed up with a Service Provider. The size of your data and the limits of your network to the outside world will have you asking the question “How I can speed up the first offsite backup to reduce the time, the cost, and the bandwidth consumption when I am sending, or syncing, my backup data to my offsite repository?”
As you know, the first backup is the big one, it is all your data; that first full copy is where the challenge resides in this scenario.
The Process – Customer Side
The process will start by backing up your data locally. You then must copy the backed up files to an external hard drive storage, or a tape, so that you can ship it to your off-site repository. The best approach to this task is to use the Veeam File Copy option from the Veeam Console:
Click the icon, and when the File Copy Wizard appears, name the copy job. Browse to your local backup repository, and then choose the backup you wish to send to the off-site repository:
Note: I chose a full backup of my SQL server from my local repository; mine is called “vRepository”.
Now that you have chosen your backup from your local source repository, you must copy this backup to a medium that you can ship to the offsite repository location.
I will copy the full backup file of my SQL server to a folder I called “C:\Off-site”. I can then ship this copy to my cloud connect server, “VeeamSrv.VeeamSaLab.org”
Note: To ensure that you do not stress your PROD environment, backup infrastructure and network, schedule the copy process to start at any time suitable for your business. You can see below that I chose to run the File Copy job at 10:00 PM.
Now that I have finished the preparation, I send the backup media to the off-site location, or to my Service Provider, and manually copy the backup files to the target repository. This process is the same method we described above, but in reverse, from media directory and folder to repository server.
Now, there is usually time while you wait for your Service Provider to copy the backup files to your remote repository.
At the Off-site Location – Service Provider
At the off-site location, you, or your Service Provider, will start the reverse process by copying the backup from the media to the repository. On the below screenshot, I added a NAS storage where I copied the “C:\Off-site” folder to and shipped to the off-site location:
At the destination step, chose the Tenant folder on your repository; in my example, the Tenant name is vUser01
Now you have chosen your folder, you can schedule when the copy to the repository must start; or, you can tick the check the box to select start after you close the File Copy wizard.
On completion of the backup file load to the Service Providers server, they will call to ask that you create a Job/copy job. Following the steps below:
Customer site – Backup Job configuration
The Tenant must scan the cloud repository first, before creating the backup, or the backup copy job.
Please note: It is recommended that a backup copy job is created and kept at the local site before sending any data to the offsite repository. This provides for a speedy recovery if anything goes wrong.
After the Rescan process has completed, you can proceed with creating a new backup, or Backup Copy Job, starting with the Backup Job you created at the start of this process, or go straight to the VM you wish to protect:
…and then Map the backup from your offsite backup repository, see below:
…and continue with your Job configuration.
In my demonstration here, the first run (copy Backup Job), a Full copy, took 22 seconds and returned the report that the “Latest restore Point is already copied”. This indicates that the seeding is working:
What about Replication to the Service Provider?
In the steps so far, I have discussed Seeding for backups. What if you or your customer wish to use Replication? Can I also use seeding with Replication, using these same steps?
The answer is yes, the process is very simple; all you need to do is to follow the steps below:
Note: I will use the same VM to replicate to the Service Provider. As the backup already exists at my remote repository, on the Service Provider side, the process will be easy and straightforward.
Create a Replication Job and choose, Low connection bandwidth (enable replica seeding):
Chose the same VM you just backed up and sent to the offsite repository; on the seeding step, map the replica as shown below:
Note: the remote Repository and replica mapping are the same from the backup process.
After you have completed the mapping and the replication job configuration, and when the replication process starts, you will notice a [seed] step on your replication job; check the screenshot following. This indicates that Veeam Replication will first copy the VM from the remote repository to the remote Datastore (locally at the service Provider side) and then start the replication process, which will be replicating the changes only.
Note: Ignore the warnings shown here; these warning not related to the actual job, but related to an SP resource configuration.
Conclusion
Whether you are a customer who wishes to apply an offsite backup strategy, or a Service Provider offering off-site Backup and DR services, bandwidth shouldn’t slow you down when you are trying to implementing you backups strategy.
From the steps above, I have demonstrated that Veeam provides great and simple steps to help you seed your backup data to an off-site repository; in return, it will save you time, bandwidth and money.
