Forum Discussion
Dataverse Virtual table link to Fabric
Hi,
I created a table in fabric workspace , with a GUID and a key
All key, GUID are unique in my table
Now in PowerPlateforme DataVerse i created a virtual table to my Fabric Table
I can see my table and datas fine.
Now i try to read my virtual table in dataverse in a flow (use dataverse read line tool)
my virtual table have more than 250 lines (select count(*) my table in fabric give seam lines number)
but in my flow the dataverse read line tool see only 27 lines
i don't understand where is the limit
thx for help
Hi ShaunLeMouton ,
Thank you for reaching out to Microsoft Fabric Community Forum, Below are the few debug points which might can resolve your issue.
- Verify that the Primary Key used in the Virtual Table is unique, contains no NULL or duplicate values, and is correctly mapped.
- Ensure the GUID is in a valid GUID format and is supported by the Dataverse Virtual Table provider.
- Confirm the Virtual Table configuration (Primary Key, Primary Name, and table mapping) is correct and that no filters are applied.
- Compare the record count between the Fabric SQL Endpoint, Dataverse Virtual Table, and Dataverse Web API to identify where the records are being filtered.
- Create a small test table and a new Virtual Table to determine whether the issue is specific to the current table or its data.
- Collect the Fabric table schema, sample data, Virtual Table configuration, and Dataverse Web API response for further investigation.
- Since the Dataverse API returns only 27 records, the issue is most likely related to the Dataverse Virtual Table/provider configuration rather than Microsoft Fabric.
Regards,
Chaithanya.
5 Replies
HI ,
27 is not a Dataverse paging boundary. The default page size for the Dataverse List rows action is 5000. So the truncation is happening below Dataverse, either in the virtual table provider or at the Fabric SQL analytics endpoint.
Run these checks in order.
- Query the SQL analytics endpoint, not a notebook. A Spark
SELECT COUNT(*) reads the delta log directly. The SQL analytics endpoint reads a synced metadata copy, and sync lag of minutes to hours is normal. Open the SQL analytics endpoint in Fabric and run the count there. If you also get 27, the problem is Fabric metadata sync, and Dataverse is innocent. - Bypass Power Automate entirely. Hit the Web API in a browser:
https://yourorg.crm4.dynamics.com/api/data/v9.2/new_yourtables. Count what comes back. 250+ rows means the flow action is at fault. 27 rows means the virtual table provider is at fault. This single test splits the problem in half, so do it before anything else. - Enable pagination in the List rows action. Settings, Pagination on, threshold 5000. Some virtual table providers advertise a small page size, and with pagination off you get one page and nothing more.
- Look for nulls and failed type conversions. Virtual table rows with a null primary key are dropped without an error. Same for rows failing conversion: dates before 1753-01-01, decimals over the declared precision, strings longer than the declared max length. 27 survivors out of 250 fits this pattern.
- Confirm your key column is a real
uniqueidentifier in Fabric, not varchar. String GUIDs parse inconsistently through the provider. - Re-read the virtual table's column mappings and data source record for a leftover filter or row limit.
Thank you!
Proud to be a Super User!
š·ļø Need more help?
ā Donāt forget to Accept as Solution if this guidance worked for you.
ā¤ļøYour Like motivates me to keep helpingHi,
I make my sql in sql terminaison end point in fabric , the select give me the correct number of line, i have checked all my key , they are correct , no null or duplicate key.
In Dataverse the API link to Datas show me only 27 entries
My key is a string , and have a GUID (string) for dataverse , (guid seem to be required as key)
- Query the SQL analytics endpoint, not a notebook. A Spark
Hi ShaunLeMouton ,
Thank you for reaching out to Microsoft Fabric Community Forum, Below are the few debug points which might can resolve your issue.
- Verify that the Primary Key used in the Virtual Table is unique, contains no NULL or duplicate values, and is correctly mapped.
- Ensure the GUID is in a valid GUID format and is supported by the Dataverse Virtual Table provider.
- Confirm the Virtual Table configuration (Primary Key, Primary Name, and table mapping) is correct and that no filters are applied.
- Compare the record count between the Fabric SQL Endpoint, Dataverse Virtual Table, and Dataverse Web API to identify where the records are being filtered.
- Create a small test table and a new Virtual Table to determine whether the issue is specific to the current table or its data.
- Collect the Fabric table schema, sample data, Virtual Table configuration, and Dataverse Web API response for further investigation.
- Since the Dataverse API returns only 27 records, the issue is most likely related to the Dataverse Virtual Table/provider configuration rather than Microsoft Fabric.
Regards,
Chaithanya.
Thankyou MJParikh for Addressing the issue.
Hi ShaunLeMouton ,Thank you for reaching out to Microsoft Fabric Community Forum,
As we havenāt heard back from you, we wanted to kindly follow up to check if the solution provided for the issue worked? or Let us know if you need any further assistance?
Regards,
Chaithanya
Thankyou @MJParikh for Addressing the issue.
Hi @ShaunLeMouton ,Thank you for reaching out to Microsoft Fabric Community Forum,
As we havenāt heard back from you, we wanted to kindly follow up to check if the solution provided for the issue worked? or Let us know if you need any further assistance?
Regards,
Chaithanya
