I had a new client for whom I thought Dataverse might be a perfect solution to their need. So I set out to give it a whirl and see if Dataverse could do the job, see if the hype and demo I had been shown by Microsoft lived up to their ‘promise’. The proverbial ‘kick the tires’ as they say.
The Database
I decide to export the database comprised of 25 tables to Dataverse. The overall schema is very straightforward, nothing too fancy here. Just mainly a series of related tables: clients, submissions, materials, trucks, …
My Experience
Disabled Dataverse Command
My first hurdle was the fact that the Dataverse buttons (both via the ribbon or via the context-menu) were disabled.
I tried updating my installation, … but that made no difference.
I also checked the official support page & and associated demonstration support video:
but neither had any information regarding this issue.
In the end, the issue is with linked tables. To export to Dataverse, you MUST use local tables (thank you Shane from the Dev Team for enlightening me on this)! So if you are trying to migrate a properly split database, the Dataverse command will not be enabled (in the front-end) as all the tables are linked. You must first delete all the linked tables and then import them into the front-end to have the command button enabled to be able to perform the export. Too bad they don’t state this in the article and save people some time and frustrations.
Why in God’s name would you disable the command button, leave it enabled and provide a message to explain to the user that tables must be local. Instead, Microsoft chose to leave users scratching their heads not knowing why the command wasn’t available. I truly don’t get the choices made by the Dev Team at times (okay, in most cases!). So, so, soooooo many opportunities missed. This is the difference between great and mediocre. This is also the difference between proper QA (with novice users) and not.
Heck, why can’t the use linked tables in the first place! The definition can be read, so there truly is no good reason it can’t be done. Even the table renaming could be done via VBA. Once again the difference between great and mediocre.
Some will say, just perform the migration in the back-end, but we want the linked Dataverse table in our front-end, not the back-end. So, this approach (doing the migration in the back-end) makes no sense in that context.
And Then The Problems Started
Finally, I was ready to perform the export. I was making progress and finally moving forward, or so I thought …
I pressed the Dataverse button, selected the tables and started the process and this is what I experienced during the process, over and over, and over again!
Let me be 100% clear here, this is the only process using the endpoint! So it is impossible that other users are performing any operations beyond the current Access instance! So once again, we see error messages that are completely incorrect/misleading!
After waiting an hour for it to bomb along all the tables, I was finally presented with:
I decided to try the ‘Create export report’, hoping to get more insight into whatever the underlying issue was, but sadly, the button did absolutely nothing. Nothing was ever generated. Another dud in my experience with Access’ Dataverse integration.
What a great choice of a caption for a button. What a great user experience.
I hate to be a broken record, but once again this is the difference between great and mediocre.
One nice thing about this Dataverse Export Summary Table, and that’s the name it should have been given!, is that is provides a URL/link to directly access the table via the Dataverse web UI so you can manage it. This avoid running in circle through the PowerApps portal trying to find things on your own!
When I logged into Dataverse I could see 3 of 25 tables had been created. None had imported any records though and the structures were not 100% correct. For instance, the Primary Key was not as defined in the Access database. So basically, not only did things fail, but what was done and left was garbage. Now I have to figure out how to delete this.
Conclusion
What more is there to say exactly, IT DOESN’T WORK! I spent hours, configuring, setting up, waiting for the process to run to end up with 3 tables imported and their structure doesn’t match the original structure (incorrect PK columns) and not a single record was imported.
I will give it another whirl today (once I figure out how to delete what was created yesterday), otherwise I will be looking into alternatives AWS, MySQL, PostgreSQL, … (so I may still revise this article should things actually work when I make another attempt at this)
Talk about frustrating and disappointing!
Imagine a non-tech savvy individual trying this out for the first time and this is/was their experience! This is supposed to sell me on the product so I promote it to my clients?!
What a disaster! I tried, I really did, but holly crap Microsoft! What the hell is going on?
The reality is I was exploring this new feature (Dataverse) for a brand new client to see if it lived up to all the hype that I had heard from people like Michael Aldridge and MVPs.
Since I was going to attempt to use this feature for the first time, I figured this would be a great opportunity to take screenshots to showcase the process and create a step by step guide, here on my blog.
So originally, I was planning on promoting the product, this article was going to be a guide for others. Sadly, the reality turned out to be completely different.
The process is very straightforward, click a button, select a few tables, click and let Access do its thing, but sadly it didn’t worked and erred. I didn’t invent the issues as I have no control over any of it. All I did why try to upsize my client’s database to Dataverse. I didn’t do anything beyond that and all I’ve done is document my experience. So don’t blame the messenger! None of it is my fault!!! I didn’t create the errors, I didn’t make Dataverse incapable of importing my client’s database, … Blame me all you want, but we all know none of that is on me.
I know there are those that hate me for it. They want to pretend everything is perfect with Access, with the Access Dev Team, … Sorry, but I tell the truth. I relay my opinion based on my experiences. Instead of blaming me they should redirect their anger and comments to Microsoft as they are solely responsible for my experience described above. All I did was try and use their tool.
It is shameful that any user would experience what I did!
More To This Story
There’s more to this story, as I continued to plug away and eventually managed to partially import some of my clients tables. Learn more about that in my 2nd part to this article:
Additional Resource(s)
Updated 2023-05-09 @10:45
I just found the following support page. So if you’re looking to venturing into Dataverse, take a look at:
Thanks DP, I rely on people like you to get in there and tell me the “real-deal”. Also, as a retired sailor, your profanity is pretty tame. I would be willing to help you in that regard as a sort of quid-pro-quo for all the help you have given me!
It’s funny, because I’ve heard other say Dataverse was great. Yet, when I try it, it completely bombed! Like I stated to the Dev Team when I forwarded all of this to them yesterday evening, is this normal or am I just lucky to hit all these problems?!
As for the profanity, if only you had been in my office for the 5-6 hours I was working on this in vain. I would have made you proud!!!
This is the major reason why I don’t jump on new features / products as soon as they’re released. Let someone ELSE swear at their computer for a few hours…
But it was supposedly beta tested for months and has been release for months, so I thought it was safe to try in a limited fashion.
I was hoping it could solve an immediate need for a client while I redevelop their solution as a proper web app using PHP. Just needed to buy a couple months time. I’ve spent a LOT of time just to get 5 table into Dataverse, and I’d have another 20 to fix and migrate and then have to adjust all the queries, forms, and reports … Sadly their db is poorly done and Dataverse doesn’t seem to manage non-ideal structures. In this specific case, their tables are all using text fields as primary keys (not my creation), once I create proper autonumber PKs in each table the imports worked. long story short the cost to make all the necessary changes, to massage their db so Dataverse is happy enough to import them, won’t be worth it to them. It is surprising as the db has worked, as is, for years in Access, yet Dataverse can’t handle it.
It does appear to work smoothy once you manage to get the table(s) into the damn thing though!
Hi Daniel, have you tried importing Access tables from within Power Apps (Home -> Tables -> Import -> Import data -> Access (data source)? I’m trialing Power Apps and Dataverse but my trial account doesn’t have Office licenses so I can’t try to do what you did.
I did try importing my Winhost SQL Server tables from within Power Apps ((Home -> Tables -> Import -> Import data -> SQL Server database), and it went pretty smoothly. I imported a 2000-row tables with a text field as PK, and Power Apps accepted it with no problem.
Import screen:
https://i.imgur.com/vyypB8I.jpg
Table data view during import; PK is “CUST_CODE”, which is text:
https://i.imgur.com/CMOWTKf.jpg
Import result; table “Customer_Billing_Addresses” successfully imported:
https://i.imgur.com/EUgY4hi.jpg