[feature request] SureBackup for the in-between people

[tomh@kuecker.com]

Feature request . . . SureBackup for the in-between people (or an alternative) - It would seem SureBackup out of box is only good for a very simple setup i.e. one esxi host with either 1 or more networks utilizing "Basic Single-host" or "Advanced single-host". But as soon as you add a second esxi host you can no longer use this out of box. Once you have a second esxi host you need to use "Advanced multi-host" which requires VMware’s “VMware Distributed vSwitch (DVS) technology” which in turn requires VMware's Enterprise Plus. The Enterprise Plus is for datacenters and cloud computing and as a small company it would be a hard sell to upgrade our Essentials license just for DVS.

Even if it is not possible to get SureBackup with multiple ESXi hosts without DVS to work, another alternate feature request would be the creation of a new job type - a restore job - in which you can create a job to restore certain files to test the ability to restore from your backup chain.

I had another backup software before I switched to Veeam, where I discovered, upon recovering a file that my backup data was corrupt even though the backup job ran without errors. After that, every week I would open each backup and restore a file. In keeping with that, when I switched to Veeam I continue to do the same thing and automating this would be fantastic.

[Egor Yakovlev]

Hi Thomas,

Thanks for feedback, noted both ideas for future improvements list.
On a side note, be aware, that recovering a single file(or a thousands) does not guarantee entire backup file consistency.

/Thanks!

[tomh@kuecker.com]

I do realize that recovering a single file does not guarantee entire backup file consistency . . . thanks for mentioning that.
Question: For those of you out there that are not able to use SureBackup, what do you do to verify backup file consistency?

[Mildur]

You can do manual instant vm recovery to a disconnected vswitch in vmware.
You can test the vm and it‘s services in the vcenter console.
After that, stop the vm instant recovery session.

https://helpcenter.veeam.com/docs/backu ... ml?ver=110

Beside disaster recovery matters, Instant VM Recovery can also be used for testing purposes. Instead of extracting VM images to production storage to perform regular disaster recovery (DR) testing, you can run a VM directly from the backup file, boot it and make sure the guest OS and applications are functioning properly.

[tomh@kuecker.com]

Thanks for the feedback I plan to do add this to my weekly checks. SureBackup does a CRC check on backup files - is performing a "Storage-level corruption guard" in advanced settings the same thing? If not, is there another method to perform a crc check?

[Mildur]

Yes, have a look at the Veeam Validator:
https://www.veeam.com/kb2086

The „Storage-level corruption guar“ checks the last restore Point for corruption, if I‘m correct.

[tomh@kuecker.com]

I was unfamiliar with Veeam Validator - thanks for pointing this tool out. I ran a sample off of one my backups and works great!

[tomh@kuecker.com]

I have been experimenting with SureBackup and discovered I can use it in a limited capacity but ran into in issue - My domain controller/dns/dhcp vm is in the same backup job as the other vm's I want to test. If I add my domain controller vm to the applications group and then add the backup job with this same vm in it to the sure backup I couldn't find a way to not start the second copy of the vm sof that it doesn't start and conflict with each other. Suggestions? Should I break out my domain controller vm into a separate backup from from the rest? I am fine with that but if I don't have to that would be great.

[Mildur]

You can create a surebackup Job for your application group and a different surebackup Job for the Backup Job (without a application Group)

Schedule them on different times.

[tomh@kuecker.com]

The only problem I see with that method is that when I run the surebackup job without an application group, I would not have a domain controller present to test ping and possible scripts. Of course if all I wanted to do is crc check then I wouldn't nee the application group.

[Taran]

In "Basic Single-host" and "Advanced single-host" the Virtual lab runs on a single VMware host. This means that the proxy appliance that creates the isolated environment and all the VMs from the application group and linked jobs are run on this host. It does not matter which hosts they were backed up from.

The idea isn't to power up all the VMs at once, rather power on the application group to provide the necessary foundation (AD...) then test the VMs from the linked jobs a few at a time. This is why the "Process simultaneously up to … VMs" field exists.

Example: The application group contains 1 Domain controller VM, the linked jobs contain 30 VMs to be tested. We've specified that we can process 3 VMs simultaneously. This means that on this host during the run of a SureBackup job we would see at most 5 VMs powered on for SureBackup on this host at any single point in time. (proxy appliance + domain controller + 3 VMs from the linked jobs.) The proxy appliance and the domain controller would be running the whole time, the rest of the VMs will cycle through 3 at a time, starting, being tested, then shutdown until all VMs have been tested.

If the host is tight on resources, reduce the number of VMs to be tested simultaneously or look at creating a virtual lab on a host with more resources.

For the most part the "Advanced Multi-host" is meant for testing replicated VMs where the replica VMs are sitting on multiple hosts.

Note: if a VM is in both an application group and in a linked job, only one copy of the VM will be started in the virtual lab. (the application group settings will override the linked job settings.) It should not be starting 2 copies of the same VM, nor should you need to make any specific exclusions. If you actually see 2 copies of the VM being started, please raise a support case.

[tomh@kuecker.com]

Ok your note addressed my concern - the issue of having the same vm start twice. So for those reading this let me recap - I have a single backup job backing multiple VM's, one of which is my Domain Controller. This VM that is my DC has been added to the Applications group. According to the note, when I fire off the SureBackup job the DC in the Applications Group is started followed by the VM's listed in the Linked Jobs, skipping the VM in the Linked Job already started by the Applications Group. Once again thanks for feedback!