Date: Mon, 9 Aug 2004 13:52:35 -0700 (PDT) From: gorgar gorgar <gorgargorgar@yahoo.com> To: freebsd-amd64@FreeBSD.org Subject: Fwd: Re: [ISC-Bugs #10805] BIND 8 & 9 segfault on AMD64 Message-ID: <20040809205235.80025.qmail@web52401.mail.yahoo.com>
next in thread | raw e-mail | index | archive | help
--0-732734451-1092084755=:78183 Content-Type: text/plain; charset=us-ascii Content-Id: Content-Disposition: inline Note: forwarded message attached. __________________________________ Do you Yahoo!? New and Improved Yahoo! Mail - Send 10MB messages! http://promotions.yahoo.com/new_mail --0-732734451-1092084755=:78183 Content-Type: message/rfc822 X-Apparently-To: gorgargorgar@yahoo.com via 206.190.39.109; Wed, 21 Jul 2004 10:49:57 -0700 X-Originating-IP: [206.190.39.112] Return-Path: <gorgargorgar@yahoo.com> Received: from 206.190.39.112 (HELO web52404.mail.yahoo.com) (206.190.39.112) by mta238.mail.scd.yahoo.com with SMTP; Wed, 21 Jul 2004 10:49:57 -0700 Received: from [216.86.204.109] by web52404.mail.yahoo.com via HTTP; Wed, 21 Jul 2004 10:49:54 PDT Date: Wed, 21 Jul 2004 10:49:54 -0700 (PDT) From: gorgar gorgar <gorgargorgar@yahoo.com> Subject: Re: [ISC-Bugs #10805] BIND 8 & 9 segfault on AMD64 To: bind8-bugs@isc.org Cc: gorgargorgar@yahoo.com In-Reply-To: <rt-10805-29807.12.0597764661484@isc.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Length: 2246 Hello there ISC! =] It's been a long time since I was able to reply to this e-mail.. so sorry for the delay. alrighty, I have backtrace info for you to look at.. it doesn't look like much, but it appears that named tried to reach a memory location that it didn't have access to and named freaked out then segfaulted. --- SNIP --- [root@server] /usr/local/bin# ./gdb GNU gdb 6.1.1 Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "amd64-unknown-freebsd5.2.1". (gdb) file /var/chroot/named/named Reading symbols from /var/chroot/named/named...done. (gdb) core ~/corefiles/named.core Core was generated by `named'. Program terminated with signal 11, Segmentation fault. #0 0x0000000000fd6000 in ?? () (gdb) bt full #0 0x0000000000fd6000 in ?? () No symbol table info available. Cannot access memory at address 0x7ffffffe9f98 (gdb) --- SNIP --- I've experienced the segfaulting with bind-9.2.3, bind-8.4.5-RC5 & bind-8.4.4, the above gdb output came from bind-8.4.4. the symptoms with bind-9.2.3 were a bit more severe.. bind-9.2.3 would segfault atleast 16 times a day if not more.. where bind-8.4.x would only segfault on average 3 times a day. So I rolled back from bind-9.2.3 to bind-8.4.x in hopes that the problem would disappear. Here is some output from 'uname' indicating our distribution and architecture type. --- SNIP --- [root@server] ~# uname -a FreeBSD server.unknown.net 5.2.1-RC2 FreeBSD 5.2.1-RC2 #1: Tue Feb 10 18:23:40 PST 2004 root@server.unknown.net:/usr/obj/usr/src/sys/SERVER amd64 --- SNIP --- if there is any further information that might be needed to help shed some light, I'll be more than happy to assist! Thank you very much, ISC! - Jeff --- via RT <bind8-bugs@isc.org> wrote: > > > > Howdy there ISC! > > > > Just reciently.. I decided to make a move towards > upgrading one of my servers > > .. I have a dual AMD64 processor machine, running > FreeBSD 5.2.1-RC2. > > > > First I tried bind9 within a chroot'd jail taking > advantage of being able to > > use dual processors, but I would get an insane > number of segfaults, luckly da > > emontools's supervise would restart named without > a problem. The problem star > > ted off small, as the days came and passed, named > would segfault even more.. > > I'd say around 10 times a day.. I decided to roll > back to bind 8.4.4 and it a > > ppeard to be fixed for the first couple of hours, > till the following day.. wh > > en I got a couple of segfaults. It seems as if > every few hours, named would s > > egfault and have to rely on "supervise" to reload > itself. > > > > I know since AMD64 is fairly new, I didn't think > running it on FreeBSD 5.2.1- > > RC2 would lead to a problem like this.. all my > other services appear to be ru > > nning fine. > > > > If you have any solutions, if you could so kindly > direct me towards some info > > rmation on how to fix this, if indeed a fix has > been released.. If you need a > > ny further information or if I could assist you in > any way about this, I'll b > > e more than happy to help!! > > > > Thanks a bunch! > > > > Jeff > > > > Do you Yahoo!? > > Yahoo! Finance Tax Center - File online. File on > time. > > There could be any number of problems here. You > are running > on a new cpu, which means the compilers havn't had > time to > be debugged fully, on a new OS, lots of untested > library > code. > > To go further we are going to need to see stack > backtraces from > the core dumps. > > Please open a new bug report for BIND9 > (bind9-bugs@isc.org). We > will keep this one for the BIND8 problems. > > Note we don't have a test machine in this > configuration. > > Mark > -- > Mark Andrews, ISC > 1 Seymour St., Dundas Valley, NSW 2117, Australia > PHONE: +61 2 9871 4742 INTERNET: > Mark_Andrews@isc.org > __________________________________ Do you Yahoo!? Vote for the stars of Yahoo!'s next ad campaign! http://advision.webevents.yahoo.com/yahoo/votelifeengine/ --0-732734451-1092084755=:78183--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040809205235.80025.qmail>