V10 RTM | Capacity Tier age out not removing from local storage

[tsukraw]

Hey guys,

Got another question with SOBR and Capacity Tier on V10 RTM if this is a expected result.

Have SOBR setup to copy backups immediately to Azure Blob. This is working perfect.
However the option "Move backups to object storage as they age out of the operational restore window" does not seem to function as I think it should. (Since V10 documentation is not out yet cant confirm)

I tried setting age out to zero days to simulate if i wanted it to only retain in Blob storage. I let it sit for 10hrs (If i recall correctly there is a process that runs every 6hrs to clean up) but the files remained on the local storage.

I adjusted to 1 day and files older then 1 day still remained on the local storage.

Question: Do these two settings "Copy immediately" and "Move on age out" not work as expected if both are enabled at the same time?

The use case i am thinking is say i only want to keep 2 days with of data in Azure Blob (I know its cheap but lets just say for DR).... And i want to keep 2 weeks on-prem.

I have my On-Prem storage do a backup to a local Repo.
Then do a copy job retaining 2 backups to a SOBR with Capacity Tiering with age out of zero so it copies that data up and clears the local Repo.

Thoughts?

[anthonyspiteri79]

Hey there.

The absolutely work together and designed as such.

My first question would be if the chain has been sealed that you want moved off and offloaded resulting in the dehydrated VBK on the Performance Tier? The offload job runs every four hours, but can be triggered manually through UI and PowerShell.

Have a read of this, and check that the chain meets the condition (effectively has a Full/Active Full run) https://helpcenter.veeam.com/docs/backu ... l?ver=95u4

[tsukraw]

Just to clarify by saying "Sealed" you mean an Inactive chain correct?
When using a Copy Job in V10 does this statement still apply for the offload?
"only full backup files that have a GFS flag on them will be offloaded"

[anthonyspiteri79]

Yes, one that is no longer under transformation.

For COPY, as soon as jobs finishes writing backup files to disk they will be copied to the Capacity Tier immediately. So using Per-VM option in a job, as it goes and completes each VM it will start the copy... even as the job is still running.

[tsukraw]

With the Copy option does it still require a GFS or could you age it out on with a zero day and right after the copy job runs each time?

The idea here is a customer who wants to use Azure as like a secondary "DR" storage. But they don't want to keep days worth of backups in Azure maybe 2-3 days (assuming they have enough bandwidth to push fulls every day up)

Now that i think about this, we would be chewing through a heck of a lot of bandwidth to accomplish this configuration. Unless Azure Blob storage and Veeam are smart enough together to track blocks and not have to re-transmit the full data file.

[anthonyspiteri79]

Hey there Tucker.

It's probably worth you having a read of the following posts and taking a look at the Cloud Field Day and Tech Field Day videos... all your questions will be answered

https://anthonyspiteri.net/tfd20-follow ... -glossary/
https://www.veeam.com/blog/v10-sneak-pe ... -mode.html
https://anthonyspiteri.net/cloud-tier-d ... on-demand/