Dataverse Backup ?

[NativeNass_]

Hi all

We were wondering if anybody has any experience with Dataverse Backups?

We are well aware about the OOB environment backup functionalities but this is not what we are looking for.

I'm thinking of data backup with multiple versions, easy restore of specific backup instead of the whole environment, ...

Thank you very much for the input

[ChrisPiasecki]

Hi @NativeNass_,

Synapse Link for Dataverse would allow you to capture incremental data changes that could allow for some type of piecemeal restore, but the restore would be a manual procedure (or you'd have to build something yourself to accomplish this).

You may want to explore a 3rd party backup solution such as OwnBackup. Their solution is meant to address the limitations described with the all or nothing OOB backup/restore. I've seen a product demo of it and is quite promising. I can put you in contact with the person I know there if you were interested in a demo, feel free to DM me. 

---
Please click Accept as Solution if my post answered your question. This will help others find solutions to similar questions. If you like my post and/or find it helpful, please consider giving it a Thumbs Up.

[FLMike]

Hello,

Just to clarify, when it says environment, it means the database itself. What do you want to not backup in the database?

Just want to help point you to something. As  @ChrisPiasecki  showed there are internal and many 3rd party that will do the "data", but restoring the data without the setting etc at times can be... well it can be a pain depending on what you want to do with it, but if you plan to restore it to the "Tenant"/"some environment", using the OOB will let you restore it to the same or other.

Cheers
If you like my answer, please Mark it as Resolved, and give it a thumbs up, so it can help others
Thank You
Michael Gernaey MCT | MCSE | MCP | Contractor Ex-Microsoft
https://gernaeysoftware.com
LinkedIn: https://www.linkedin.com/in/michaelgernaey

[NativeNass_]

Hi @FLMike 

Well as far as I know, when you restore the environment if just one person made a mistake, you lose all your progress because it resets the complete environment to a certain point in time?

This is from my perspective not a valid solution to present to a customer who is using Dataverse to a great extent.

We are aware of some third party tooling but I was wondering if there are also native solutions possible.

Thanks for your response already

[NativeNass_]

hi @ChrisPiasecki 

Thanks for your response. We have also been finding third party solutions for backup but were wondering if there is anything possible that is part of the Microsoft Ecosystem without the need for a third-party tool.

[FLMike]

Hi,

It really depends on what exactly is broken. Let me explain. While it is true some tables are less connected than others, in many cases, heck lots of them, changes go into many tables. Its difficult to rectify changes across a large number of tables, plugin runs, etc. So yes, the restore environment is sort of the big bang affect I guess lol :-), but when you think about backing things up, just make sure to thing in terms of business process, lookups, etc, which isn't always easy to know unless you wrote it all.

Definitely agree, sometimes you just wanna fix some records 🙂 lol and Lord knows you dont wanna do it based off change tracking or logs all the time.. uggg

Wish I could recommend something for you but realistically its a crap shoot.

[NativeNass_]

@FLMike 

Thanks for the respons!

Yea indeed a lot of dependencies but hey, since they created the platform I think they might have thought of a valid backup solution that doesn't require a big bang effect like it does today 🙂

It is indeed a whole process when you need to restore record values manually based on data storage that was actually meant for analytics purposes. Let alone explain this in a proper way to a client.

Guess there is no other way than going third party 🙂

[FLMike]

Actually I am responding lol how I am as I was on the product team for many many years and decades at MS using the product and there unfortunately has never been a great solution.

I wish I could say there was but, pretty much turning on change tracking for whatever you want and then using Synapse or running ADF jobs etc, which require a lot of pagination blah blah..

Its why so many people wanted to be able to treat it like a stand alone SQL instance. Made the onprem easy for people..cloud not so much.

[FLMike]

Truly just a big part of the issue is its not just a SQL server. Between the plugins that exist on the platform, custom ones you make, workflows that trigger, automate flow that trigger etc, restoring data can be very painful, especially if you forgot to disable those while you load, or forget to turn them back on. annd the list goes on.

Great as long as you dont have an issue. after that its a dart board sometimes and you hope you hit it

[NativeNass_]

@FLMike 

Indeed a lot of dependencies that come into picture when you start thinking about it.

On the other hand, Clients don't care and just want to have something that works 😄

We'll just try out some different scenario's and see what fits the best.

Probably start messing around with the Data Lake Exports/Synapse and see how that goes in terms of recovery scenario's.

Thanks for all the effort 🙂