no calculated mean in .data files

After comparing carefully with old and new data, I have found another problem: I downloaded the data for a third time, and without changing anything in the analyse-window, a few participants who had a calculated mean this morning, now also had missing calculated means ? I am at an complete loss as I did not make any changes. Can this be a bug?

Also, somewhat related: In the beginning after I got live I added a save BLOCKLIST command and specified in the analyse format to exclude certain trials based on blocklist info. I understand that for the first few participants (which did not had any blocklist info) these trials can not be excluded and the reaction time is not correct. Yet the total N of the trials in those participants suggests it still excluded those trials? I checked the means by calculating them by hand and they are indeed not correct. What is excluded here? I have already collected and downloaded data since adding and changing the analysis-criteria, and the calculated means for the 'new' participants worked fine untill today.