Date: Sat, 31 Aug 2002 23:17:45 +0200 From: Arnvid Karstad <arnvid@karstad.org> To: Matthew Dillon <dillon@apollo.backplane.com>, bmah@FreeBSD.ORG, freebsd-stable@FreeBSD.ORG Subject: Re: Problems with FreeBSD - causing zalloc to return 0 ?! Message-ID: <20020831231356.8BCC.ARNVID@karstad.org> In-Reply-To: <200208312106.g7VL6JEU002376@apollo.backplane.com> References: <20020830190849.8B8A.ARNVID@karstad.org> <200208312106.g7VL6JEU002376@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 31 Aug 2002 14:06:19 -0700 (PDT) - Matthew Dillon <dillon@apollo.backplane.com> wrote: > Ok, I've examined the kernel core dump. I'm still not sure why > INVARIANTS made any difference, but the kernel is definitely running > out of KVM and it looks like the main culprit is the number of > mbufs and mbuf clusters configured. It looks like they were > manually configured up. let's see now root@irc:/usr/local/etc/rc.d# more /etc/sysctl.conf kern.maxfiles=65536 kern.maxfilesperproc=65536 net.inet.udp.blackhole=1 net.inet.tcp.blackhole=1 #kern.maxproc=1024 kern.maxprocperuid=1024 net.inet.ip.fw.dyn_max=10000 root@irc:/usr/local/etc/rc.d# more /boot/loader.conf # -- sysinstall generated deltas -- # userconfig_script_load="YES" kern.ipc.nmbclusters=129536 net.inet.tcp.tcbhashsize=65536 and relevant kernel configs: options NMBCLUSTERS=32384 # Up clusters options DFLDSIZ=(512*1024*1024) #Increase options ICMP_BANDLIM All of these were up'd in cooperation with someone with greater knowledge of the inner issues of FreeBSD than myself after we experienced problems with nic lock downs etc related to the load of appx 20000 users at the same time.. Mvh/Best regards, Arnvid L. Karstad To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020831231356.8BCC.ARNVID>