Is it possible that file-rename is not an atomic operation?? It appears that data destined for the old file name is somehow finding its way into the renamed file after the rename has supposedly been executed. Does anyone know anything about this?? --Robyn