Showing results for 
Search instead for 
Did you mean: 
New Member

Data Flow Publish Fails

I am trying to create a simple data flow to pull some SP List information into CDS so I can present the data via a Portal App.  When I step through the Data Import wizard everything works as expected.  However, when I check the status of the flow under Data Flows, it always shows that publishing failed.  When I check the refresh history all I see is:


I don't see any option to "republish" (how does one do that?).  
I've tried using manual, as well as automated (scheduled), publishing, but neither one works.  What do I have to do to get this flow to actually publish?  Is this just a system delay and it will eventually retry automatically, or do I have to do something different?
Super User
Super User

Hello @rtmystic 


You are not having an option called Republish directly. You need to click on THREE dots, click Edit, make some changes in the QUERY then Click Next --> Check the Table too --> Click Next --> and PUBILSH it again.


Its like a kind of altering, if you dont have any issue, recreate and retrigger for a safer side.



Could you please tel me which field you are trying to update. Is it a lookup field?

Because in your screenshot am not seeing any field getting mapped


Please mark as Answer if it is helpful and provide Kudos

Subscribe :
Blog :


  • I'm not trying to update "any field" per se ... I just want to do a first time sync and use the dataflow to get my SP List Items into CDS ... What I did was:
    DataVerse > Tables > Data > Get Data
  • I selected SP Online List
  • Created my connection
  • Selected the list I wanted
  • Removed some of the "System Columns" (I did not want/need them)
  • Tried both Manual and Scheduled sync settings
  • Saved the configurations, which then created the dataflow and it showed that it was being "published" 
  • Eventually the Dataverse > Dataflows screen updated to show that publishing had failed

When I try to republish as you suggest I do this:

  • Go to Dataverse > Dataflows and select my flow


  • Click [...] > Edit


  • To make some changes to the query, I removed the 'Id' column (which the query wizard added), since I could use the default SP ID column as the unique identifier
  • Clicked Next
  • Updated the Mapping Settings to load a new table (since it did not exist yet) and made the ID column the primary column, then clicked Publish 


  • Under Dataverse > Dataflows it shows publishing is in progress, and eventually it shows it failed again



  • When I look at Refresh History, all it shows is an error occurred ... no details







Advocate II
Advocate II

@rtmystic I'm having the exact same issue. The error I'm getting after the publish fails is useless. All it says is "Something went wrong" and in the details it gives a session and a request GUID. 

Were you able to get around this? if so, what was your solution? 

I submitted a ticket to Microsoft support. Hopefully I can find out what the issue is. 


do you get any clue from Microsoft?

Advocate II
Advocate II


I discovered that my alternate keys weren't properly defined on the table I was importing. I had to redefine my alternate keys and redo the import. Then it worked. It's been a few weeks now. So I can't remember exactly the key that was missing. But here's the info that MSFT added to the ticket closure information. Hope it helps:


MSFT REPLY: I want to inform you that I shared your case with my technical team for review, and after a discussion, I want to share with you our outcomes:
Lookup columns create relationships between Dataverse tables, utilizing a specific GUID recognized only by Dataverse. You can find more information about lookup columns here: Create a relationship between tables by using a lookup column - Power Apps | Microsoft Docs
To map a Dataverse record (GUID) using lookup columns to an external system like PowerApps dataflows, the "Alternate Keys" must be defined as stated in the following Microsoft document: Define alternate keys to reference rows with Microsoft Dataverse - Power Apps | Microsoft Docs. It states:
"Alternate keys provide an efficient and accurate way of integrating data with external systems. It’s essential in cases when an external system doesn’t store the Globally Unique Identifier (GUID) IDs that uniquely identify rows in Microsoft Dataverse."
I also want to share other links to helpful information for this process:
- Define alternate keys using Power Apps - Power Apps | Microsoft Docs
- How to map a Lookup Column in a Power Platform Dataflow (
The lack of alternate keys was the reason of these errors, but this information should address the process of creating a data flow using lookup columns for you in the future.

Helpful resources

Power Apps News & Annoucements carousel

Power Apps News & Announcements

Keep up to date with current events and community announcements in the Power Apps community.

Microsoft 365 Conference – December 6-8, 2022

Microsoft 365 Conference – December 6-8, 2022

Join us in Las Vegas to experience community, incredible learning opportunities, and connections that will help grow skills, know-how, and more.

Power Apps Community Blog Carousel

Power Apps Community Blog

Check out the latest Community Blog from the community!

Users online (3,857)