Date: Mon, 26 Jan 2004 14:21:56 +0100 From: Andre Oppermann <andre@freebsd.org> To: sebastian ssmoller <sebastian.ssmoller@gmx.net> Cc: current@freebsd.org Subject: Re: panic: m_copm, length > size of mbuf chain Message-ID: <40151474.8010406@freebsd.org> In-Reply-To: <20040125203456.7bfaa67b.sebastian.ssmoller@gmx.net> References: <20040125203456.7bfaa67b.sebastian.ssmoller@gmx.net>
next in thread | previous in thread | raw e-mail | index | archive | help
sebastian ssmoller wrote: > hi, > my system ran stable for several days now but suddenly this error occured. At this moment i did nothing special - just surfing the net ... > > cause i have no serial line attached i could only get the following infos: > Debugger(...) at Debugger+0x54 > panic(...) at panic+0xd5 > m_copym(...) > ip_fragment(...) > ip_output(...) > ip_forward(...) > > i didnt want to write down all these numbers(addresses) by hand, should i ? > > it seems to have to do something with inet/ip so i possibly should mention that i disabled "option inet6" in kernel and that i use "pf" as > packet filter. > > $ uname -a > FreeBSD tyrael.linnet 5.2-CURRENT FreeBSD 5.2-CURRENT #2: Mon Jan 19 22:26:43 CET 2004 root@tyrael.linnet:/usr/obj/usr/src-current/sys/DEBUG i386 > > i can provide more info if needed... This looks like a host/network byte order problem in relation with the 'pf' packet filter. ip_output() expects the ip_off field to be in host byte order. If this is not the case you will get bogus length values for fragmentation. -- Andre
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?40151474.8010406>