Honestly, I'm slightly concerned because that doesn't sound right. Do you have a workflow that we can use to reproduce this?
I checked the code. The node would fail if this error is printed. So I wonder why your parallel loop completes (does it show the green traffic light and the virtual nodes disappear again?).
Now it seems I cannot reproduce this problem but I can tell you it was linked to fact I tried to execute this internal parallelized loop inside a partially executed external loop. If I remember correctly the problem disappeared after I manually deleted the virtual metanode that is automatically created when the parallel loop is executed and then I restarted the external loop from the beginning. I hope this can help if other people run into this problem.
I will write here more details if the problem occurs again.
OK, thanks. So that was caused by some parallel looper that was half-way executed and then started without a proper reset before? Yes, that should be fixed and we should better warn if these semi-executed loops are resumed (in fact we do for partially executed loops that get loaded from disk).
... that's one of the reasons why the parallel loop construct is still in labs. You need to clean-up yourself if it fails.
Dear All,
this was a helpful chat. I had the same issue on the KNIME server. After I deleted the virtual node prior to uploading the workflow to the server, the error message disappeared.
Thanks
Stefan