PISA Errors are classifed two categories. One is "BIMPCT" and the other is "ZEBRA error". Those were reported at PHENIX computing meeting on Janurary 2000 by Y.Watanabe. C.Maguire commented that it was due toFULUKA and his simulation group was going to patch this problem by skipping bad event. There are two kind of "ZEBRA error". Large fraction of this error was occured at perticular node "ap14". It is suspected that the machine itself has a problem. A few runs were tested if it was reproducable. Failed jobs at ap14 were not crush again. But certainly, few run reproduced "ZEBRA error" at the same event regardless which node was used. This seems to be software problem.