> > What if you try with the latest POM and kernel (20051027 and 2.6.14 > resp.) ? > > I'm not a programmer so if this won't work, I don't know. I also find > it strange that you do have a .so, but not a .ko file. > > > Gr, > Rob Thanks anyway. I will try it, but i Don't think it can be work. Justnow I have read an article, It paste parts of the source, but i found it is diffrent with that in my machine. I don't know which is the 'right'. but that's code not the latest, it maybe may or jul's. here's the code of him if (0 == memcmp(&conn->tuple,&tuple,sizeof(tuple)) && found != NULL && (found_ct = tuplehash_to_ctrack(found)) != NULL && found_ct->proto.tcp.state != TCP_CONNTRACK_TIME_WAIT) { anzhen then is in my machine if (0 == memcmp(&conn->tuple,&tuple,sizeof(tuple)) && found != NULL && found->proto.tcp.state != TCP_CONNTRACK_TIME_WAIT) { you can found the diffrent with them. I think it is big. but I don't know which will be right. And He said he's code was not right too, should be modify to these: if (found != NULL && (found_ct = tuplehash_to_ctrack(found)) != NULL && 0 == memcmp(&conn->tuple,&tuple,sizeof(tuple)) && found_ct->proto.tcp.state != TCP_CONNTRACK_TIME_WAIT) { or else it will be let the linux crazy, and the errors were ( <0>Kernel panic - not syncing: Fatal exception in interrupt) ^_^