From owner-freebsd-net@FreeBSD.ORG Sun Jun 26 00:17:34 2011 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 644F21065670 for ; Sun, 26 Jun 2011 00:17:34 +0000 (UTC) (envelope-from dudu@dudu.ro) Received: from mail-qy0-f175.google.com (mail-qy0-f175.google.com [209.85.216.175]) by mx1.freebsd.org (Postfix) with ESMTP id 2DF9E8FC0A for ; Sun, 26 Jun 2011 00:17:33 +0000 (UTC) Received: by qyk30 with SMTP id 30so951975qyk.13 for ; Sat, 25 Jun 2011 17:17:33 -0700 (PDT) Received: by 10.229.134.199 with SMTP id k7mr3562084qct.171.1309045785195; Sat, 25 Jun 2011 16:49:45 -0700 (PDT) MIME-Version: 1.0 Received: by 10.229.39.210 with HTTP; Sat, 25 Jun 2011 16:49:05 -0700 (PDT) In-Reply-To: References: From: Vlad Galu Date: Sun, 26 Jun 2011 01:49:05 +0200 Message-ID: To: Adrian Minta Content-Type: text/plain; charset=ISO-8859-1 X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Cc: freebsd-net@freebsd.org Subject: Re: FreeBSD 8.2 and MPD5 stability issues X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 26 Jun 2011 00:17:34 -0000 On Sat, Jun 25, 2011 at 11:28 PM, Adrian Minta wrote: > After recompilation with "*default release=cvs tag=RELENG_8" and pooling > disables the system still crashes around 4200 sessions. The server has a > xeon E5520 CPU and 4G of ram. Here is the crash on the screen: > http://img232.imageshack.us/img232/6751/crashm.png > > $uname -a > FreeBSD lns 8.2-STABLE FreeBSD 8.2-STABLE #2: Sat Jun 25 23:50:06 EEST > 2011 gygy@lns:/usr/obj/usr/src/sys/MYKERNELv3 amd64 > > /boot/loader.conf: > > net.graph.maxdata=65536 > net.graph.maxalloc=65536 > kern.ipc.maxpipekva=320000000 > #I wave 4 cores > net.isr.bindthreads=4 > net.isr.maxthreads=1 > net.isr.defaultqlimit=4096 > > /etc/sysctl.conf: > > net.inet.ip.forwarding=1 > kern.maxfiles=65535 > net.inet.ip.intr_queue_maxlen=8192 > net.inet.icmp.icmplim=0 > kern.ipc.nmbclusters=1280000 > kern.ipc.maxsockbuf=128000000 > net.graph.maxdgram=10240000 > net.graph.recvspace=10240000 > kern.random.sys.harvest.ethernet=0 > net.isr.direct=1 > net.isr.direct_force=0 > > > Kernel GENERIC plus: > > #options DEVICE_POLLING > options HZ=1000 > options NETGRAPH > options NETGRAPH_PPPOE > options NETGRAPH_SOCKET > options NETGRAPH_CISCO > options NETGRAPH_ECHO > options NETGRAPH_FRAME_RELAY > options NETGRAPH_HOLE > options NETGRAPH_KSOCKET > options NETGRAPH_LMI > options NETGRAPH_RFC1490 > options NETGRAPH_TTY > options NETGRAPH_ASYNC > options NETGRAPH_BPF > options NETGRAPH_ETHER > options NETGRAPH_IFACE > options NETGRAPH_L2TP > options NETGRAPH_MPPC_ENCRYPTION > options NETGRAPH_PPP > options NETGRAPH_PPTPGRE > options NETGRAPH_TEE > options NETGRAPH_UI > options NETGRAPH_VJC > > options ALTQ > options ALTQ_CBQ > options ALTQ_RED > options ALTQ_RIO > options ALTQ_HFSC > options ALTQ_PRIQ > # options ALTQ_NOPCC > > device pf > device pflog > device pfsync > device gre > > device tap > device hme > > # Bridging > device if_bridge > device lagg > > options IPSTEALTH > options ZERO_COPY_SOCKETS > > options NETGRAPH_DEBUG > options GDB > options DDB > options KDB_UNATTENDED > > > Hi Adrian, You will need to set dumpdev to something (e.g. "AUTO") in /etc/rc.conf (it defaults to NO) so that your kernel coredump is saved to your swap partition before rebooting. Also, you might want to remove ZERO_COPY_SOCKETS. -- Good, fast & cheap. Pick any two.