Veeam Jobs with VM Tags

[dspringer]

May I ask you about your swarm knowledge? I'll put the question directly here, even if it's indirectly about vSphere and could belong in a subforum.

External service providers keep recommending that I convert my backup jobs to VM tags. But from my point of view, this can only be implemented to a limited extent - unless I create lots of VM tags.

As things stand now:
For example, I have 2 jobs for an external site in which I back up file servers and the rest separately. All servers are entered there manually. File servers have only indexing active in “Guest Processing”. The application servers, on the other hand, have this deactivated and have “application-aware processing” active instead. For domain controllers, I have VSS active as “Require success”. I also need a special user for the Active Directory.

Planned:
I have created 4 different tags: File server, application server, archive, no backup.
The file server is easily implemented and tested.
But what do I do with the application servers? I have them all in the corresponding VM tag - no problem. But on the Veeam side, I can't simply differentiate between the two within the backup job: This server runs with standard credentials and this server requires a special login. Also, “application-aware processing” can only be set directly for the selected tag - not for the objects it contains.

Do you use VM tagging and how would you proceed in this case? Using this example: An extra VM tag just for the domain controller so that I can integrate it separately?

[DChiavari]

Hi David, if I got your question right, I think you can accomplish this quite easily. It might not be very intuitive, but you can actually differentiate credentials, guest processing and indexing options for individual items (and/or logical or physical containres) inside a job.

By clicking on the "Applications...", "Indexing..." and "Credentials..." buttons you can add overrides to the general job settings, that would normally be applied by default to all items processed by the job.



In my small example here, I created a job adding a datastore as its selection list. That datastore has 2 VMs inside, one being a Windows Domain Controller and the other an Ubuntu server. I disabled Application-Aware processing for the Linux VM.



In a similar fashion, I disabled Indexing for the Windows VM.



Of course instead of specific VMs, you can use tags (or tags combinations) for overrides. Hope this helps!