Home > Veeam B&R > Veeam: Backup & Recovery v7.0 – Basic Setup

Veeam: Backup & Recovery v7.0 – Basic Setup

In my previous post, I covered the installation of Veeam B&R v7.0.  Now I’ll review the basic setup

1) Launch the program via the shortcut.

image

2) Click on the BACKUP INFRASTRUCTURE button.

image

Click on ADD SERVER and then choose VMWARE VSPHERE.

3) Enter the FQDN of your vCenter Server.

image

4) Enter credentials for the vCenter Server.

image

I create a service account, “svcVeeamAdmin” that has rights both to vCenter and Domain Admins for internal VM tasks such as VSS, Indexing, etc.  Follow whatever best practices apply for your environment.

5) On the summary screen, click FINISH.

image

6) If you expand Managed Servers -> VMware vSphere -> vCenter Servers -> <vCenter Server>, you should see your cluster hosts on the right.

image

7) Click Backup Repositories and then click ADD REPOSITORY

image

Name the repository appropriately so you can easily identify it later.

8) Choose the type of Backup Repository.  While Microsoft Windows Server and Shared Folder (CIFS/SMB) may seem similar, the Windows Server option allows for the installation of an agent to help with efficiency and a Shared Folder would be more like a Linux server or NAS that is exposing a CIFS share but is not otherwise running Windows.

image

9) Enter the FQDN of your repository server:

image

10) Choose the credentials you added previously:

image

11) When you click NEXT, it will check for the presence of an agent on the destination, and then allow you to install or upgrade.

image

12) Once the installation completes, you’ll see the summary.

image

Click Next

13) On the summary screen, click FINISH to complete creating the repository.

image

14) Click POPULATE to show the drives on the system with the agent installed.

image

Select the drive you want to use, and click Next.

15) Click BROWSE or type in the PATH TO FOLDER, then click POPULATE.

image

You can also control the load – the number of concurrent backups or limit the MB/sec (so as to not saturate the network).

If you click ADVANCED:

image

You get options to align backup blocks and decompress the backups – largely for when writing to a repository that supports its own deduplication.  Different use cases will have the remote path obtaining better results than the Veeam deduplication.

16) Next, we configure the vPower NFS folder.

image

We can select to enable it – or not.  Also you can select a Write Cache folder – such as an SSD path, to speed up the writes.  Click Next.

17) On the REVIEW page, you’ll see a summary:

image

Click NEXT

18) And then on the APPLY screen, click FINISH.

image

 

19) Click on the Backup & Replication button, and then click Backup Job.

image

Name your backup job, and click Next.

20) Click ADD to add VM’s to your backup:

image

21) The default view, when you expand the tree shows your Hosts & Clusters view, with your VM’s – just like the vCenter Client.

image

I’ll pick my running VM’s, but an alternative might be to use a VM’s & Template’s view, and pick VM folders.

image

22) Next, we need to pick a Backup Proxy (Used to speed up backups if the Veeam B&R VM cannot attach to them locally) and Backup Repository (to store the backup files).  The Backup Repository can be off host, such as a DAS or a NAS that is not part of your Shared Storage – remember not to keep your backups where your primary data is!

image

Select the REPOSITORY we create earlier.

Note the “Configure Secondary Destination for this job” – this will let you store a second copy on a second repository.  We’ll cover that another day, after we make our first backup job.

23) Click the ADVANCED button to get access to the good options:

image

I personally prefer “Reverse Incremental” which makes the previous night’s backup file the Full.  This does require a higher IOPS rate, and there are reasons why other options are better.

I would recommend clicking the check box for PERFORM ACTIVE FULL BACKUPS PERIODICALLY if you want to create a new FULL backup every now and again.  It’s up to you if you wish to choose “forever incremental” or have occasional full’s.  Its really a matter of confidence.

24) Click on the STORAGE tab:

image

Here you set the Deduplication, Compression, and optimize the storage method.

25) Click on the NOTIFICATIONS tab:

image

Put in an e-mail address to send your alerts to.  I recommend checking the VM NOTES box and choose a field other than “NOTES”.  If you use NOTES the default NOTES field in vCenter Client will get overwritten – and you might be using that for something.  Veeam is nice enough to let you pick the field you want to use, and if it does not exist, it will create it.

26) Click on the vSphere tab:

image

Here, you choose if you want to use VMware Tools quiescing – this will get you “Application Consistent” backups vs “Crash Consistent”.  Not using CBT (Change Block Tracking), which only backs up new/changed blocks from the previous backups, would be just silly.

27) Click on the ADVANCED tab:

image

The defaults are pretty good, but look into the manual to decide if you want to fiddle with this.

28) The STORAGE INTEGRATION tab:

image

Will allow you to use the SAN snapshot to do the backup.  I understand this is limited to just the HP VSA currently, but let’s the SAN snapshots be used, with the assumption that the SAN tools created clean, application consistent quiesced backups.

Click OK.

29) You’ll get a warning that you need to set up e-mail options:

image

So we’ll go do that in a bit.

30) Back on the NEW BACKUP JOB screen, we go to the GUEST PROCESSING option:

image

If you’re just looking for the VM’s and crash-consistent backups, then you don’t have to change anything.  There are many reasons why you might want this.  However, I would check the boxes for Enable application-aware image processing (for VSS to truncate things like Exchange and SQL logs) and then Enable Guest File system indexing (if you want to search backups for particular files).  Choose the credentials we created earlier:

image

If you click the ADVANCED button, you can modify the previous options on a per-VM basis.

Click NEXT.

31) On the SCHEDULE window:

image

Set up your schedule to run on particular days and times.  Veeam has an excellent Automatic Retry option where it will retry failed VM’s at the end of the job.  Often this is enough to bypass whatever made the VM fail in the first place – maybe it was being backed up by something else, it was too busy to quiesce, etc.  I’ve seen other backup products fail an entire job of 50 VM’s if only a small error occurred and not retry – or retry the WHOLE job.

You also have the option to automatically terminate the job if it exceeds the backup window – such as 8AM in the morning, when staff start showing up.

Click CREATE!

32) On the summary screen, click RUN THE JOB if you wish, and click FINISH.

image

33) Now let the job run.

image

34) Let’s go configure those mail options for backup notifications.  Click the upper left hand corner icon and choose OPTIONS:

image

35) On the E-Mail Settings tab:

image

Check the ENABLE box and enter your SMTP server name, and TO/FROM accounts.  Click TEST MESSAGE and verify it works, then click OK.

 

That’s really about it.  I realize at 35 steps, it *seems* like a lot, but that’s not just creating the first job, but setting up the entire system, the repository, the proxy, the notifications, etc.  They’re also very easy steps.  At this point you have a fully functioning daily backup of your VM infrastructure.

 

When you click on the job, it will show you details in the lower pane.  Depending on available resources (number of CPU’s), Veeam will choose the number of parallel VM’s it can back up.  More vCPU’s = More simultaneous VM’s.

image

Here you can see that I’m getting 81.9MB – a decent rate considering I only have 1GbE link to my media server.  More interesting is that it has processed 27.5GB, but only had to transfer 13.8GB – a 2x savings in bandwidth.  You can see the deduplication and compression savings live.  This rate tends to get increasingly better as additional VM’s are processed, as they likely contain very similar data if they’re all similar OS’s with similar roles or built from similar templates.

image

And now that the job is done (A different job, so the times are off), you can see that my 8 VM’s took 93 minutes to backup, the bottleneck was actually reading from the source (1GbE NFS, who knew!), processed 325GB, but only transferred 79.2GB – a 3.9x dedupe ratio.

image

An incremental, run immediately after shows a 38 minute duration that read 245GB of VM’s, only had to read 48.9GB of actual data, and only transferred 1.3GB of it due to deduplication and compression.  Imagine what that would be like across the WAN compared to a ‘normal’ backup!

I’ll have more posts on the new Veeam B&R v7.0 coming, but head on over to our local Veeam SE, Rick Byrne for more advanced options – http://rickrbyrne.wordpress.com/

Categories: Veeam B&R
  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: