Although the old connection string works for pipelines, I changed some of the parameters & their order and it works for Data flows now too:
User ID=
vs.
User=
added:
Trusted_Connection=False;
and
ApplicationIntent=ReadOnly;Persist Security Info=True
moved to the end;
Dynamic dataset not working in Data flows but does work in Pipeline - Help!
I have an Azure Data Factory Pipeline that uses Connection services with parameterizable properties that passes the table schema name for an Azure SQL DB.
It has a Managed Virtual Network Integration runtime with Interactive authoring disabled.
And it works... But...
It does not seem to work with Mapping Data flows
I believe the connection string uses SQL Authentication.
IP Whitelisting is used necessitating the Managed Virtual Network.
Is there a workaround or configuration where this would work?
I'm experimenting now, but welcome some insight on things that simply aren't supported or workarounds that make things work.
I should also say that its also using Azure Key Vault as well. The Secret name is dynamic.
This is the error message I get when I test the connection:
Spark job failed: { "text/plain": "{\"runId\":\"[SomeGUID]\",\"sessionId\":\"[SomeGUID]\",\"status\":\"Failed\",\"payload\":{\"statusCode\":400,\"shortMessage\":\"com.microsoft.dataflow.broker.InvalidOperationException: Only one valid authentication should be used for [Some Name]. SQLAuthentication is invalid. One of user/password is missing.\",\"detailedMessage\":\"Failure 2022-07-06 18:55:43.499 failed DebugManager.processJob, run=[SomeGUID], errorMessage=com.microsoft.dataflow.broker.InvalidOperationException: Only one valid authentication should be used for [Some Name]. SQLAuthentication is invalid. One of user/password is missing.\"}}\n" } - RunId:[Some GUID]
1 answer
Sort by: Most helpful
-
Scott Stauffer 11 Reputation points
2022-07-06T21:36:53.917+00:00