Forum Discussion
Unable to load DLL 'ParquetSharpNative':
boley2131 tonyalexanderIM StuartTurner , the latest gateway version (3000.222.10) has been published and it contains the fix for the pipeline side as well. Please try it out.
Is there a known issue/ recommendation for VMWare machines ? We have installed the latest Gateway version but are still getting this error
StuartTurner the issue around AVX2 instruction has been resolved, so you are probably running into a new issue.
Can you share your GW version and the exact error message?
GW version 3000.214.8
ErrorCode=UserErrorFailedFileOperation,'Type=Microsoft.DataTransfer.Common.Shared.HybridDeliveryException,Message=The file operation is failed, upload file failed at path: 'c64e7d4a-1a16-4b4d-b35a-77733bd38772/9a784df0-744c-41fb-8fa8-94e8291cdd48/Tables/PeriodTest'.,Source=mscorlib,''Type=System.TypeInitializationException,Message=The type initializer for 'Microsoft.DataTransfer.ClientLibrary.ParquetV2WriteTypeFactory' threw an exception.,Source=Microsoft.DataTransfer.Richfile.ParquetTransferPlugin,''Type=System.DllNotFoundException,Message=Unable to load DLL 'ParquetSharpNative': A dynamic link library (DLL) initialization routine failed. (Exception from HRESULT: 0x8007045A),Source=ParquetSharp,'
OK, that is the March release but coming from Pipelines. Since we have combined ADF SHIR and the Power BI Gateway, the Parquet problem needs to be patched in both sides.
Are you using Pipelines directly or are you trying out the fast copy via pipelines feature in Dataflows?
We're using Pipelines directly - Copy data activity. To ingest data from an on-premise SQL Server database.
Dataflows are working (haven't tried fast copy, my undestanding is this is only available for Azure SQL). However, we are implementing a data driven load pattern using Pipelines.
Is there anything in the April release that addresses this issue?
Not sure what you mean by patching both sides, we can only update the Gateway - is there something that can be applied in Fabric?
Thanks