Date: Sun, 1 Oct 2000 13:43:50 -0500 (CDT) From: BSD <bsd@shell-server.com> To: stable@freebsd.org Subject: Constant panics: some thoughts on the situation Message-ID: <Pine.BSF.4.10.10010011309320.70294-100000@marvin.shell-server.com>
next in thread | raw e-mail | index | archive | help
Let me start off by saying I'm no kernel hacker. So whatever I say here may be totally wrong and I would appreciate being corrected. Anyways, I've noticed some things about this whole situation. So here's a list of observations. 1) It seems to be affecting computers that have a relatively high network traffic load. I, for one, get hit with every attack imaginable. I'm sure Andrew J.Caines of altavista.net gets tons of funky packets too. 2) When in tcp_timer.c, tcp_timer_persist() function, it gets passed 0xdfd10720 as its void *xtp argument. It then sets struct tcpcb *tp = xtp. At no point in the function do I see tp being changed (and it's never passed with &tp)...other than that hack which calls a goto out;. But from the crash dump, the trap is on line 340, which is before the out label. When printing tp, the value shows as 0x620000. How did this change? 3) And why did this change to THAT #. Michael Allman's crash dump shows his trap coming in at route.c:135/rtalloc1(). When he prints the value of rnh, he gets 0x62000000. 2 extra 0s. Is this a mere coincidence? Is there something special about this # in the BSD kernel? --Bart To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.10.10010011309320.70294-100000>