What you have pointed out is the intended behaviour of ArcRORB. Once ArcRORB has exported the catchment file, the catchment file then needs to be loaded into the RORB GE and then saved through the RORB GE.
After saving through the RORB GE, you should see that some changes have been made to the catchment file. The version number of the RORB GE that you have used will be there, the column of nodes should be populated (that column you pointed out represents the downstream node) and the rest of the catchment file should be filled out. The text that ArcRORB outputs is only used as part of the RORB GE to visualise the catchment file, after saving through the RORB GE the rest catchment file will then be filled in based on what you have outputted from ArcRORB.