Error updating the semantic model from the Power BI service

[jafp]

From the Power BI service, when refreshing a semantic model that connects to an AWS EMR via the Microsoft Hive ODBC driver the following error occurs:

Error

Data source error: {"error":{"code":"DM_GWPipeline_Gateway_MashupDataAccessError","pbi.error":{"code":"DM_GWPipeline_Gateway_MashupDataAccessError","parameters":{},"details":[{"code":"DM_ErrorDetailNameCode_UnderlyingErrorCode","detail":{"type":1,"value":"-2147467259"}},{"code":"DM_ErrorDetailNameCode_UnderlyingErrorMessage","detail":{"type":1,"value":"ERROR [08S01] [Microsoft][Hardy] (115) Connection failed with error: [Microsoft][Hardy] (79) Failed to reconnect to server."}},{"code":"DM_ErrorDetailNameCode_UnderlyingHResult","detail":{"type":1,"value":"-2147467259"}},{"code":"Microsoft.Data.Mashup.ValueError.Reason","detail":{"type":1,"value":"DataSource.Error"}}],"exceptionCulprit":1}}}

If we refresh from Power BI Desktop, the semantic model refreshes correctly. We have updated the on-premises data gateway to the latest version but the same error persists. In the semantic model we have incremental refresh configured like other models that do not error although those have less data volume. In any case they do not exceed 1 GB.
Thanks you very much.

[v-echaithra]

Hi @jafp ,

May I ask if you have resolved this issue? Please let us know if you have any further issues, we are happy to help.

Thank you.

[v-echaithra]

Hi @jafp ,

We’d like to follow up regarding the recent concern. Kindly confirm whether the issue has been resolved, or if further assistance is still required. We are available to support you and are committed to helping you reach a resolution.

Best Regards,
Chaithra E.

[v-echaithra]

Hi @jafp ,

I hope the information provided is helpful.I wanted to check whether you were able to resolve the issue with the provided solutions.Please let us know if you need any further assistance.

Thank you.

[jafp]

Good afternoon,
I have tried disabling the incremental refresh policy on each of the tables, keeping the filter with the RangeStart and RangeEnd parameters, and when updating from the Power BI service, it does not give an error.

[v-echaithra · Accepted solution]

Hi @jafp,

Thanks for your response @GilbertQ .

Sometimes date range passing can absolutely be the root cause if the parameters are misapplied or mismatched with the source system’s expectations.
Make sure the query filters on the date column using the parameters explicitly, Check that DateColumn matches exactly the date field in your Hive table.
Check date formats passed to Hive. You can add a diagnostic step in Power Query to output the SQL query sent via ODBC (sometimes via tracing or logs). Alternatively, test the raw query with parameters directly in Hive or a SQL client to confirm it returns the expected results with the correct date filtering.
Verify Hive table partitioning, if the table is partitioned on the date column common for big data tables, your filter should match that partition column exactly for optimal performance. If not partitioned, even a filter won’t save you from scanning large data volumes.
Temporarily replace RangeStart and RangeEnd with fixed dates you know exist in the source, then test refresh from the service to confirm if the filtering works properly and the refresh succeeds.

Hope this helps.
Warm Regards
Chaithra E.

[GilbertQ]

So it could be an issue with the way the dates in the date range are getting passed to the source system?

[v-echaithra]

Hi @jafp ,

Compare Desktop vs. Service Mashup Queries

Run the same query from Desktop (on gateway server), then from Service. In Power BI Desktop:
Open Query Diagnostics > “Start Diagnostics”
Check for Query duration, Number of rows, Transformation complexity. The Power BI Service has max execution time (default ~600 seconds, but often less), lower resource allowances compared to local Desktop. If the query approaches service limits, the gateway will disconnect or timeout causing the exact [08S01] and reconnect failures you’re seeing.

Look at the logs on your on-premises data gateway under: C:\Program Files\On-premises data gateway\GatewayLogs\
Filter for MashupDataAccessError and see if any timeout or memory-related entries exist.

Split Incremental Loads (Partitioning)

If the dataset is large:
Use smaller incremental refresh windows (e.g., daily instead of monthly).
Implement RangeStart and RangeEnd filters in your query to reduce in-memory size.

Check the gateway machine:
Ensure at least 8 GB RAM, ideally more.
Check CPU/RAM usage during refreshes.
Consider installing gateway on a dedicated VM for heavy workloads.

Create a test dataset:
Use the same ODBC connection but point to a smaller table or filter rows.
Try refreshing it from Power BI Service.
If it works: issue is likely data volume, transformation complexity, or timeout.

Hope this helps.
Warm Regards
Chaithra E

[GilbertQ]

Hi @jafp 

What happens if you log in to your gateway server, open up the Power BI desktop file and try and refresh? What happens then? The reason for me asking you to do this is to see if there is a connectivity issue between your gateway and your data source?

[jafp]

Good morning @GilbertQ,
From the server where the gateway is installed, the update works from Power BI Desktop. This is how we can update it without any errors. The problem occurs only if we update it from the Power BI service.
Thank you very much.

[FBergamaschi]

There is an issue with the datasource. The error can be interpreted and used to fix the problem but you need to open a support ticket to Microsoft

go to

https://admin.powerplatform.microsoft.com/

and create a ticket. They will immediately support you

If this helped, please consider giving kudos and mark as a solution

@me in replies or I'll lose your thread

Want to check your DAX skills? Answer my biweekly DAX challenges on the kubisco Linkedin page

Consider voting this Power BI idea

Francesco Bergamaschi

MBA, M.Eng, M.Econ, Professor of BI