Data360 Analyze

Hi, I'm looking for some clarification on connection points, I have an issue i can't quite figure out.  Full disclaimer this is a data flow that was converted from an LAE graph but the original LAE imported bundlers have been deleted and the connection points have been recreated in D360.

Here's the issue:

I have 9 data sets from a connection point coming into my composite and I want to select one of the data sets for input into the filter.  

When I hook the connection point to the filter I'm presented with the single data set by tag option and I select the "Rates" data set to feed into the filter node.

However, when I attempt to run the node it errors and returns "too many data sets provided" even though I had selected the "Rates" data set in the previous step.

Is there something I'm not understanding or setting up correctly?

I created a simple test graph in LAE to investigate the operation. Multiple data sets are presented to a Bundler and the bundle is connected to the input of a Composite node:

The Composite node then contains an Unbundler node and the output with the second data set is connected to a Filter node:

The legacy data flow was then imported into Analyze. The legacy Bundler/Unbundler nodes were then converted to Connection Points (I did not delete them and create new Connection Points), which resulted in the following:

I did not receive any errors when all the nodes were run. Perhaps you could step through the example on your machine to see what results you get. The only difference I can see from the screenshots you provided is that, in my case, the Transform (Superseded) node is connected to the output of the Connection Point within the Composite node. However, in your situation, the Transform (Superseded) node is connected directly to the Input pin of the Composite node. Perhaps you could also try moving the connection (blue arrow) to the Connection Point (red arrow) to see whether that resolves the issue: