Hello,
On October 28, 2026, an Azure Data Factory Mapping Data Flow in the Japan East region failed with the following error:
Encountered the following exception when trying to create Remote Application Folder:
System.NullReferenceException: Object reference not set to an instance of an object.
at Microsoft.Marlin.Common.Storage.Impl.DataLakeServiceClientFactory.GetDataLakeServiceClient(String traceId, String storageAccountUri, String fileSystemName, String sasToken, Boolean skipCache) ... line 102
at Microsoft.Marlin.Common.ClusterStorage.ADLSGen2ClusterStorageWithUdk.GetADLSGen2FileSystemClient(...)
at Microsoft.Marlin.Common.ClusterStorage.ADLSGen2ClusterStorageWithUdk.<>c__DisplayClass26_0.<CreateRemoteApplicationFolderAsync>b__3>d.MoveNext() ... line 503
...
at Polly.Retry.AsyncRetryEngine.ImplementationAsync[TResult](...)
at Polly.AsyncPolicy.ExecuteAsync(...)
at Microsoft.Marlin.Common.ClusterStorage.ADLSGen2ClusterStorageWithUdk.CreateRemoteApplicationFolderAsync(...) ... line 473
The Data Flow activity reported error code 145 (System error).
The Data Flow uses Azure Blob Storage with Account key authentication and a Managed Virtual Network Integration Runtime.
The Data Flow was configured to retry up to five times, but the activity still failed. When the batch was rerun later that night, it completed successfully without any changes to the configuration or input data.
Could you please help clarify the following?
Could this exception be caused by a transient issue during Mapping Data Flow runtime initialization or by an internal Azure service?
Why does the stack trace reference ADLS Gen2 when the source and sink use Azure Blob Storage?
Are there any known issues affecting Mapping Data Flow or Integration Runtime in Japan East around October 28, 2026?
What additional diagnostic information should we collect if this issue occurs again?
Thank you.