Azure Best Practices (MSP)

[infused]

Hi Guys.

Looking to answer a few of my questions around this:

1. Blob per customer? or single blob for all customers? (separate containers within the blob)
2. Backing up whole org vs individual jobs? We found the following works, but unsure
2a. https://i.imgur.com/3XRfFwT.png
2b. https://i.imgur.com/hl9I7qG.png (org backup ignores users)
3c. https://i.imgur.com/QNutP5c.png (Group backup focuses on users and mail)

The above is done to keep getting all the sites/onedrives/teams. Avoids grabbing utility accounts etc. Is this a good way of doing things?

We want to be able to stop touching the backup solution all the time. This solution allows us to add users to/from a group without bothering about sharepoint/teams - it just happens.

[Mildur]

Hi Hayden

1. Blob per customer? or single blob for all customers? (separate containers within the blob)

I would suggest different Storage Accounts per customer. If you want to bill the API Calls, egress costs and used storage to your customer, a storage account per customer will allow you generate reports dedicated to each customer. I'm not aware of any direct method in Azure to get this reports on a container level.
If you don't care about such reports or have your found an own way to export this usage data, then a single storage account will also work. But I recommend to create different containers per customer to be used as dedicated VB365 backup repositories. In my opinion it is easier to manage customers data in Azure Blob in case a customer has cancelled your service.

2. Backing up whole org vs individual jobs? We found the following works, but unsure
2a. https://i.imgur.com/3XRfFwT.png
2b. https://i.imgur.com/hl9I7qG.png (org backup ignores users)
3c. https://i.imgur.com/QNutP5c.png (Group backup focuses on users and mail)

Our recommended best practice is to separate repositories per data type and customer. Retention is managed on repository level.
For me it's important to backup SharePoint and teams to the same repository. If not, your repository would would require a lot more disk space because SharePoint and Teams Jobs could backup the same data (Teams Files).
With that in mind, individual jobs per data type are recommended. You can still configure the job to automatically add all new objects to the job. You have done that already in your screenshots: https://imgur.com/hl9I7qG

Just keep the configuration maximums per Job and Proxy in mind: https://bp.veeam.com/vb365/guide/design/maxconfig.html

My job design would look like this:
- Exchange Job (Entire Org, only Mail + Archive) in Repo "Exchange Repo"
- One Drive Job (Entire Org, only OneDrive) in Repo "SPO Repo" or separate "ODB Repo"
- Sharepoint Job (Entire Org, Sharepoint + Teams) in Repo "SPO Repo"

Thanks
Fabian

[infused]

That's kind of how we were doing our jobs, except we split teams and sharepoint, which has caused issues.

Are MSPs charging for the endpoint calls? We're just doing a flat fee per user, which isn't really working anymore.

[Mildur]

except we split teams and sharepoint, which has caused issues.

May I ask what sort of issues you are facing?

Are MSPs charging for the endpoint calls?

I know of service providers who wants todo that.
In Azure Blob, you pay for the storage, API Calls and Data Egress. If one customer restores a lot of data from object storage, it can cost you some money.

backing up email via groups does not show a restore option in the restore portal

M365 Group Mailbox or if you use a security group to backup the user mailboxes?
M365 Groups Data cannot be restored from the Restore Portal. But I don't know any limitation about user mailboxes backed up through a security group.

Thanks
Fabian

[Mildur]

I moved your comments to a new Topic for the Restore Portal question.

Thanks
Fabian