Date: Tue, 15 May 2007 19:44:52 +0200 From: Attilio Rao <attilio@FreeBSD.org> To: Stefan Bethke <stb@lassitu.de> Cc: FreeBSD Current <freebsd-current@freebsd.org>, Andre Oppermann <andre@freebsd.org>, Kris Kennaway <kris@obsecurity.org> Subject: Re: panic: mutex tcp owned at /usr/src/sys/netinet/tcp_input.c:2475 Message-ID: <4649F194.7070106@FreeBSD.org> In-Reply-To: <47AE9AC9-13D4-45A0-B199-CFB348D9B284@lassitu.de> References: <97D4D7C1-8FE4-48C1-9604-A567420796D5@lassitu.de> <4644D331.9080901@freebsd.org> <5729FCC6-7CD5-4D26-B075-9D517B9A06D5@lassitu.de> <3bbf2fe10705131245y276af14as53f3839e62024473@mail.gmail.com> <355DA9CE-F097-46EA-8D37-BA4C27019246@lassitu.de> <20070513222114.GA64807@xor.obsecurity.org> <68AA0B73-75E0-4E20-8AE0-74FB2FB654ED@lassitu.de> <20070513232413.GA65863@xor.obsecurity.org> <C8F48051-6C5A-4BAC-B226-EF1F02F9F051@lassitu.de> <20070514080510.GB71110@xor.obsecurity.org> <47AE9AC9-13D4-45A0-B199-CFB348D9B284@lassitu.de>
next in thread | previous in thread | raw e-mail | index | archive | help
Stefan Bethke wrote: > > Am 14.05.2007 um 10:05 schrieb Kris Kennaway: > >> On Mon, May 14, 2007 at 09:57:13AM +0200, Stefan Bethke wrote: >> >>> OK, got it. However, it's a bit more than I'm willing to hand- >>> transcribe :-) >>> >>> Any way of saving this info without a serial console or a digital >>> camera? >>> >>> I'll try to get a serial port the next couple of days... >> >> I think ktrdump can operate on a core. > > Sorry, the first file was truncated. For the benefit of the list, I've > uploaded it here: > http://www.lassitu.de/freebsd/20070510/ > http://www.lassitu.de/freebsd/20070510/ktrdump.txt.gz Hello Stefan, thanks a lot for your effort, I've seen exactly what I expected (a leaked recursion on 0xffffff002ae8dbd8 mutex (that I suspect is "TCP"). Since none of us can do on-line debugging in your machine, can you please download that patch, apply it, recompile your kernel and doing these tests again?: http://users.gufi.org/~rookie/works/patches/kern_mutex.diff It will give a good understanding of where the recursion happens and where probabilly expected unrecursion fails. Thanks, Attilio
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4649F194.7070106>