From owner-freebsd-stable@FreeBSD.ORG Thu Jul 22 01:58:55 2010 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 88E291065676 for ; Thu, 22 Jul 2010 01:58:55 +0000 (UTC) (envelope-from mike@sentex.net) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by mx1.freebsd.org (Postfix) with ESMTP id 48BF78FC26 for ; Thu, 22 Jul 2010 01:58:54 +0000 (UTC) Received: from mdt-xp.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.14.4/8.14.3) with ESMTP id o6M1wrNx069393 for ; Wed, 21 Jul 2010 21:58:53 -0400 (EDT) (envelope-from mike@sentex.net) Message-Id: <201007220158.o6M1wrNx069393@lava.sentex.ca> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Wed, 21 Jul 2010 21:58:59 -0400 To: freebsd-stable@freebsd.org From: Mike Tancsa Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Subject: RELENG_8 flowtable issue ? X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 22 Jul 2010 01:58:55 -0000 I noticed on a recent RELENG_8 box, the CPU became pegged. Looking at top, it has something to do with flowtables last pid: 49269; load averages: 2.76, 2.29, 2.07 up 18+09:36:04 21:42:07 89 processes: 6 running, 70 sleeping, 13 waiting CPU: 0.0% user, 0.0% nice, 50.0% system, 0.0% interrupt, 50.0% idle Mem: 52M Active, 1496M Inact, 172M Wired, 16K Cache, 112M Buf, 280M Free Swap: 3072M Total, 3072M Free PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND 10 root 2 171 ki31 0K 16K RUN 0 809.5H 100.00% idle 18 root 1 44 - 0K 8K CPU0 0 117:30 100.00% flowcleaner as the flowcleaner is running full speed. Does anyone know what that might be about ? sysctl -a |grep flow kern.sigqueue.overflow: 0 net.inet.ip.output_flowtable_size: 32768 net.inet.tcp.reass.overflows: 0 net.inet.flowtable.stats: net.inet.flowtable.nmbflows: 50176 net.inet.flowtable.tcp_expire: 86400 net.inet.flowtable.fin_wait_expire: 600 net.inet.flowtable.udp_expire: 300 net.inet.flowtable.syn_expire: 300 net.inet.flowtable.enable: 0 net.inet.flowtable.debug: 0 net.inet6.ip6.auto_flowlabel: 1 # sysctl -a net.inet.flowtable.stats net.inet.flowtable.stats: table name: ipv4 collisions: 958 allocated: 0 misses: 992897 max_depth: 1 free_checks: 36419211 frees: 785503 hits: 9745703272 lookups: 9746696168 The box runs mpd5 to terminate some l2tp connections and runs ospf and has just one area. In its releng7 incarnation, all was quite solid. I am going to disable flowtables on this box for now, but if I see it happen on another, what can I do to better debug it ? ---Mike -------------------------------------------------------------------- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, mike@sentex.net Providing Internet since 1994 www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike