V3 Questions

[Rouxls Kaard]

Hello,
1. I'm reading through the documentation for V3 and saw this blurb:

IMPORTANT
Veeam Backup for Microsoft Azure does not support archive tiering for storage accounts with enabled blob soft delete or data redundancy (ZRS, GZRS, RA-GZRS) options.

Does V3 support archive tiering for storage accounts with GRS? I'm thinking about enabling this on my storage account in order to make my data more resilient than what LRS currently offers, but if I can't restore/backup to archive tier with GRS enabled, then it's back to the drawing board.

2. What do you recommend for making the Veeam appliance fault tolerant? In my scenario, the region where my storage account and appliance are goes down, it's totally lost. But, since I had GRS enabled, my data is available in the fail-over region. I was reading that if a different appliance tries to backup or restore to data that was managed by another appliance, data-loss could occur. Should I opt for just restoring my original appliance to the fail-over region so it can continue to backup/restore? Or perhaps keep a second appliance running in the fail-over region that can backup/restore from the GRS-protected data?

Thanks!!

[micoolpaul]

Hi!

I can’t provide any definitive answer on question 1 but I did find an interesting note within Microsoft’s documentation that they themselves don’t support ZRS for archive tier:

https://docs.microsoft.com/en-us/azure/ ... redundancy

“ 1 The archive tier is not currently supported for ZRS accounts.”

If you go one step further and check the Azure calculator you can see that Azure only support LRS/GRS and RA-GRS as Archive Tier options. So that may explain Veeam’s deliberate highlighting of those unsupported tiers! https://azure.microsoft.com/en-gb/prici ... dn=disable

2. Depending on budgets you could look to leverage Veeam B&R either on-prem or in another region and copy your Azure backups that way to help work towards 3-2-1-1-0. I like having an on-prem backup copy so I haven’t got a single point of failure in the Azure platform, whether due to billing problems, compromised identity or environmental failure.

[nielsengelen]

1. Correct, we follow the same logic as Microsoft here.
2. We are looking into enhancements in the future for a quick recovery of the appliance. For now, I do believe your proposal will work as long as you only have 1 instance running. Keep in mind that even if you need to do a full install of the appliance, you can import existing backups and we can restore from those without issues as we store all relevant data within the backup restore points.