If you meant that you want to continue processing the same dataflow once error occurred then it is not possible in DS and that is very logical. If there is an error in a dataflow (no matter what kind of) then you may capture this in an error log table using the method suggested in the article above and continue the execution with rest of the dataflows.