In the course of climate change, extreme weather events and their consequences are likely to increase in the next decades. To enable publicly available predictions preceding an event, we operate an Advanced Research Weather Research & Forecasting (WRF-ARW) limited area model to ensure pro-active mitigation strategies before the start of a storm event. To demonstrate the actual model performance for multiple stakeholders, we compared the prediction with publicly available measurements from nearby stations recorded during the extreme event, starting in the evening of October 22nd, 2014 in the Mondsee catchment. In the beginning, the model prediction highly underestimated the rainfall at most of the weather stations. However, the prediction accuracy increased from 54 to 30 hours in advance to the event. For the Mondsee weather station located inside the catchment, the predictions 30 and 6 hours in advance had an accuracy of -32% and -28%, respectively. However, the prediction was challenged by extremely unstable weather conditions. Nevertheless, the prediction forecasted an event where flooding was a very likely consequence, considering the spatial-temporal amount of rainfall predicted. Thus, an early warning message to responsible stakeholders would have been appropriate for pro-active mitigation action in this case.